This action might not be possible to undo. Are you sure you want to continue?
by Robert L. Read
How to be a Programmer: A Short, Comprehensive, and Personal Summary
by Robert L. Read Published 2002 Copyright © 2002, 2003 Robert L. Read
Copyright © 2002, 2003 by Robert L. Read. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with one Invariant Section being ‘History (As of May, 2003)’, no Front-Cover Texts, and one Back-Cover Text: ‘The original version of this document was written by Robert L. Read without renumeration and dedicated to the programmers of Hire.com.’ A copy of the license is included in the section entitled ‘GNU Free Documentation License’. The home of the transparent electronic copy of this document is: http://samizdat.mines.edu. Revision History Revision 1.8 22 Apr 2003 DocBook format, GFDL, and major fixes Revision 1.5 03 Feb 2003 Incorporated Slashdot feedback and fixed typos Revision 1.0 01 Dec 2002 First publishing of pdf at Samizdat
To the programmers of Hire.com.
Table of Contents
1. Introduction ....................................................................................................... 1. ................................................................................................................4 2. Beginner ........................................................................................................... 1. Personal Skills ...........................................................................................5 1.1. Learn to Debug ................................................................................ 5 1.2. How to Debug by Splitting the Problem Space ....................................... 6 1.3. How to Remove an Error ................................................................... 7 1.4. How to Debug Using a Log ................................................................ 7 1.5. How to Understand Performance Problems ........................................... 8 1.6. How to Fix Performance Problems ...................................................... 8 1.7. How to Optimize Loops ..................................................................... 9 1.8. How to Deal with I/O Expense ............................................................ 10 1.9. How to Manage Memory ................................................................... 10 1.10. How to Deal with Intermittent Bugs ................................................... 11 1.11. How to Learn Design Skills .............................................................. 12 1.12. How to Conduct Experiments ........................................................... 12 2. Team Skills ...............................................................................................13 2.1. Why Estimation is Important .............................................................. 13 2.2. How to Estimate Programming Time ................................................... 13 2.3. How to Find Out Information ............................................................. 15 2.4. How to Utilize People as Information Sources ....................................... 15 2.5. How to Document Wisely .................................................................. 16 2.6. How to Work with Poor Code ............................................................. 17 2.7. How to Use Source Code Control ........................................................ 18 2.8. How to Unit Test .............................................................................. 18 2.9. Take Breaks when Stumped ............................................................... 18 2.10. How to Recognize When to Go Home ................................................ 18 2.11. How to Deal with Difficult People ..................................................... 19 3. Intermediate ...................................................................................................... 1. Personal Skills ...........................................................................................21 1.1. How to Stay Motivated ...................................................................... 21 1.2. How to be Widely Trusted ................................................................. 21 1.3. How to Tradeoff Time vs. Space ......................................................... 21 1.4. How to Stress Test ............................................................................ 22 1.5. How to Balance Brevity and Abstraction .............................................. 23 1.6. How to Learn New Skills ................................................................... 24 1.7. Learn to Type .................................................................................. 24 1.8. How to Do Integration Testing ............................................................ 24 1.9. Communication Languages ................................................................24 2. Team Skills ...............................................................................................25 2.1. How to Manage Development Time ..................................................... 25 2.2. How to Manage Third-Party Software Risks .......................................... 26 2.3. How to Manage Consultants ............................................................... 26 2.4. How to Communicate the Right Amount .............................................. 26 2.5. How to Disagree Honestly and Get Away with It .................................... 27 3. Judgement .................................................................................................27
......... 53 9.................8. 36 3...............................7......................................................1......................... How to Know When to Apply Fancy Computer Science .......... 31 4...............1...2............................................................ TRANSLATION .................................. 28 3.52 7............2........................................ 36 3.............. TERMINATION .................... How to Tell the Hard From the Impossible ....................................9................................................................3............................ How to Decide if Software is Too Immature .............................................................................. 30 3......................... COPYING IN QUANTITY .................................. A.................... 2003) .................... 28 3........ How to Deal with Managerial Myths ....................... 40 3....................... Comprehensive............... History (As Of May...................... 33 1.................... 1.............. How to Communicate Well .................. 38 3.. 30 3...................... Choosing Languages ... and Per- 3......................... How to Grow a System ................... How to Fight Schedule Pressure ................................................................................................... How to Evaluate Interviewees .... 27 3............. COMBINING DOCUMENTS ....... How to Get the Most From Your Teammates ...............................7.............................................6.... PREAMBLE ...................49 4..... B............. APPLICABILITY AND DEFINITIONS .......................................... AGGREGATION WITH INDEPENDENT WORKS ........ 36 3.... 37 3.............. 41 Glossary ............ How to Get a Promotion .... 34 2.... C......................................... 33 1................................. How to Manage Software System Dependence ...........................53 11................. How to Understand the User ......53 10....................................... How to Choose What to Work On ...... 53 12............................4... 39 3........................................................ VERBATIM COPYING ............. Advanced ...................1.5.............................. ADDENDUM: How to use this License for your documents ............................3....................................................................4................................ 31 3........................... How to Make a Buy vs.....................34 2....................................... 50 5.......................... Compromising Wisely . How to Handle Boring Tasks ......3................. 29 3........ How to Tradeoff Quality Against Development Time ..................6.................11....How to be a Programmer: A Short............................ 38 3...... How to Utilize Embedded Languages .......................................................... 52 8............. 54 iii .........10........................................................ How to Gather Support for a Project ............................. How to Grow Professionally ....................... MODIFICATIONS .. 39 3...................... Build Decision ... How to Divide Problems ............................2......................................1............................................................................. COLLECTIONS OF DOCUMENTS ...............2.............................34 2.......48 2........................................................................................................ Technological Judgment .................................... 37 3................................................................. 40 3.......................... How to Tell People Things They Don't Want to Hear ............. 1...................3............... 48 3................. Serving Your Team ................................................................................................. 38 3........... Bibliography ........................5..........8...............33 1......................................... 35 2........................50 6........... GNU Free Documentation License .................................... How to Deal with Temporary Organizational Chaos ..... How to Talk to Non-Engineers .......................................... How to Develop Talent .............................. FUTURE REVISIONS OF THIS LICENSE ................
This essay differs from those excellent works by emphasizing social problems and comprehensively summarizing the entire set of necessary skills as I see them. Computer programming is taught in courses. This is very subjective and. company connotes the modern workplace and tribe is generally the people you share loyalty with. 4 . company. Introduction To be a good programmer is difficult and noble. and tribe. The essays of Paul Graham[PGSite] and Eric Raymond[Hacker] should certainly be read before or along with this article. you may wish to save paper by not printing some of the appendices. therefore. The hardest part of making real a collective vision of a software project is dealing with one's coworkers and customers. Code Complete [CodeC93]. and Extreme Programming Explained [XP99] all teach computer programming and the larger issues of being a good programmer. Note If you are printing this for your personal use. I confine myself to problems that a programmer is very likely to have to face in her work. But it is really child's play compared to everything else that a good programmer must do to make a software system that succeeds for both the customer and myriad colleagues for whom she is partially responsible. Writing computer programs is important and takes great intelligence and skill. I hope in spite of this that this essay will be useful. Welcome to the tribe. The excellent books: The Pragmatic Programmer [Prag99]. Rapid Development [RDev96]. I use the words business.Chapter 1. In this essay I attempt to summarize as concisely as possible those things that I wish someone had explained to me when I was twenty-one. synonymously except that business connotes moneymaking. In this essay the term boss to refer to whomever gives you projects to do. this essay is doomed to be personal and somewhat opinionated. Many of these problems and their solutions are so general to the human condition that I will probably seem preachy.
examine the code you wrote and have no clue how the error can be occurring. Debugging tools are wonderful when they are stable and available. When it doesn't. The first meaning of the verb to debug is to remove errors. you must debug. 5 . typically adding lines that print information out. More commonly. such as checking an assertion against a large data structure.1. Most of this code is imperfect and imperfectly documented. you usually don't get to see the program. Personal Skills 1. it involves things that are not meant to be visible. this means some assumption you are making is not quite correct. that require writing code and changing the execution of the program. Learn to Debug Debugging is the cornerstone of being a programmer.Creating a permanent window into the programs execution in the form of a log. there are some kinds of debugging. But there will still arise places where the code does not conform to the documentation (crashing your entire machine is a common and spectacular example). or some condition arises that you did not anticipate. If you buy something from a major software company. or complexity theory. In addition. but it is critical to be able to employ the other two methods. or where the documentation is mute. These hidden things must be revealed. and your colleagues. or whatnot. you create an error. not programs themselves. are more fundamental are not working programmers. The common ways of looking into the ‘innards’ of an executing program can be categorized as: • • • Using a debugging tool. A programmer that cannot debug effectively is blind. but the meaning that really matters is to see into the execution of a program by examining it. Finally. or whether certain assertions hold across a complicated data structure. Debugging is about the running of programs. which lines of code are actually being executed. Sometimes the magic trick of staring into the source code works.Making a temporary modification to the program. because the debugging tool may subtly change the way the program executes it may not always be practical. Debugging tools often lag behind language development. Even if you are perfect. Without the ability to gain visibility into the execution of this code the slightest bump will throw you permanently. Inevitably. To get visibility into the execution of a program you must be able to execute the code and observe something about it. debugging. organizations like GNU. like what is being displayed on a screen. Sometimes this is visible. or analysis. like the state of some variables inside the code. so at any point in time they may not be available.Chapter 2. your are surrounded by and must interact with code written by major software companies. Beginner 1. Often this visibility can only be gained by experimentation. and Logging --. that is. In many other cases. The working programmer does not live in an ideal world. Idealists that think design. Printlining --. or the delay between two events. It is good to know how to use debugging tools when they are stable. but the printlining and logging are even more important.
But what is the middle of a mystery? There is where true creativity and experience comes in. When several people are working together under pressure on a problem it is easy to forget what the most important mystery is.Beginner Some beginners fear debugging when it requires modifying code. When out look at the output. such as the space of executed lines. The last three are not visible from the output. you have to learn to experiment on it. such as ‘It crashed on thing #9. When you run it. the memory management. so now your mystery is smaller: ‘It crashed on thing #8. Once you have evenly subdivided the space of all that can go wrong. you now have a mystery. but instead it does something else. the interaction with foreign code. You don't have the vision you will later develop to see the other dimensions of the program. and understand that nothing that you temporarily do to it will make it worse. Having that mental model is what helps one find the middle of the mystery effectively.’ I find that bearing in mind exactly what the mystery is at any point in time helps keep one focused. For the experience programmer. you must try to decide in which space the error lies. 6 . It is not always quite so simple---any examples I can give will be contrived compared to what sometimes happens in practice. these other dimensions form an imperfect but very useful mental model of all the things that can go wrong. When we run it again. You can use a debugger or we can add printline statements (or the equivalent in whatever language you are working in) after #8 and #9. our mystery will be smaller. the space of all possible errors looks like every line in the source code. #9. 1. you can ask yourself: ‘Is the unknown line executed before or after this line that I judge to be executed in the about the middle of the running program?’ Usually you will not be so lucky as to know that the error exists in a single line. This is understandable---it is a little like exploratory surgery. it crashes. for example. To a true beginner. How to Debug by Splitting the Problem Space Debugging is fun. Suppose. or even a single block. and you will be debugging quickly. seek out a mentor---we lose a lot of good programmers at the delicate onset of their learning to this fear. But you have to learn to poke at the code and make it jump. You think it should do something. If there is a single key to debugging is to use the divide and conquer technique on the mystery. If you feel this fear.2. the code that is risky. you created a program that should do ten things in a sequence. Debugging requires creativity and ingenuity. you see that the first seven things in the sequence were run successfully. because it begins with a mystery. The key to divide and conquer as a debugging technique is the same as it is for algorithm design: as long as you do a good job splitting the mystery in the middle. Since you didn't program it to crash.’ Can you design an experiment to see which thing it crashed on? Sure. Often the mystery will be more like: ‘Either there is a pointer in that graph that points to the wrong node.’ In that case you may have to write a small program to check that the pointers in the graph are all correct in order to decide which part of the subdivided mystery can be eliminated. the data structure. and the code that is simple. you won't have to split it too many times. or #10. or my algorithm that adds up the variables in that graph doesn't work. In the simple case where the mystery is: ‘Which single unknown line makes my program crash?’.
Decide what it should do. system architects must understand and use logs because of the complexity of the system. such as the value of some variable. You may have to fix a system you cannot rewrite. Ideally you will have perfect understanding of the code and will reach an ‘A-Ha!’ moment where you perfectly see the error and how to fix it. you must exercise your experience and judgment and assign your own meaning to the code. and then rerun the program and observe that the bug no longer exists. Sometimes it is unclear what the program should do or what the original author intended. but you should still conceptually apply a single atomic change to fix the bug. When configurable. making what is output configurable is very useful. logs offer three basic advantages: • • • Logs can provide useful information about bugs that are hard to reproduce (such as those that occur in the production environment but that cannot be reproduced in the test environment). For this reason. logs allow general information to be captured in order to debug unanticipated specific problems without having to modify and/or redeploy the code just to deal with those specific problems.3. there are really several bugs that look like one. each record in the log will identify its position in the source code. and comment it or clarify it in some way and then make the code conform to your meaning. Printlining is just producing a simple. ideally while the program is running. 1. you want to make the smallest change that fixes the bug. log. usually temporary. In general. You may see other things that need improvement. the thread that executed it if applicable. called a log. etc. Too little information and it may not contain what you need. But of course. and. Absolute beginners must understand and use logs because their knowledge of the programming is limited. How to Remove an Error I've intentionally separated the act of examining a program's execution from the act of fixing an error. sometimes more than one line must be changed. Attempt to employ the scientific method of changing one thing and only one thing at a time. Of course. But since your program will often use insufficiently documented systems into which you have no visibility. Too much information makes the log expensive and produces scroll blindness. but the real world is often messy. Typically. an additional useful piece of information. such as the time passing between statements. The amount of information that is provided by the log should be configurable. In other cases the code is so complicated that your understanding cannot be perfect. making it hard to find the information you need. the amount of free memory. then put your fix in place. the number of data objects. The amount to output into the log is always a compromise between information and brevity. Logs can provide statistics and data relevant to performance. It is up to you to define the bugs and fix them one at a time. but don't fix those at the same time. These log statements are 7 . Sometimes. The best process for this is to be able to easily reproduce the bug. debugging does also mean removing the bug. this is not always possible. In this case.Beginner 1. How to Debug Using a Log Logging is the practice of writing a system so that it produces a sequence of informative records. the precise time of execution. commonly.4. In fixing a bug. This is an intermediate or advanced skill that is sometimes harder than writing the original function in the first place.
You should design the log statements to address problems that you anticipate. and often it must eventually be considered more carefully. Finding the expensive I/O and the expensive 10% of the code is a good first step to building your mental model. Even if you understand perfectly and precisely the cost of the code you write. it is both wise and effective to choose a solution that gets the job done simply and quickly. and some of the debugging statements will probably be permanently added to the logging system. There is a famous dictum that 90% of the time will be spent in 10% of the code. in practice performance problems are a little different and a little easier than debugging in general. be far more expensive than processor seconds. database or other server accesses may. your code will make calls into other software systems that you have little control over or visibility into. network bandwidth. performance is a part of usability. There is not much sense in optimizing a function that accounts for only 1% of the computation time. The first resource to measure is wall--clock time. this may not always represent the whole picture. 1. Deadlock is the inability to proceed because of improper synchronization or resource demands. it is best to have a way of measuring this contention from the start of your project. 1. Each statement can be assigned a level and will only output a record if the system is currently configured to output that level. I would add to that the importance of input/output expense (I/O) to performance issues. However. Often most of the time is spent in I/O in one way or another. How to Fix Performance Problems Most software projects can be made 10 to 100 times faster than they are at the time that they are first released with relatively little effort.Beginner sprinkled throughout the source code but are particularly at major functionality points and around risky code. in the end. The key to improving the performance of a very complicated system is to analyze it well enough to find the bottlenecks. Before you try to make it faster. memory. Suppose that you or your customers consider a system or a subsystem to be too slow. As a rule of thumb you should think carefully before doing anything unless you think it is going to make the 8 . it is very helpful to be able to assert that with confidence. However. but less efficiently than some other solution. Contention for shared resources that are synchronized can cause deadlock and starvation. If it can be at all anticipated. Logging wall-clock time is particularly valuable because it can inform about unpredictable circumstance that arise in situations where other profiling is impractical. Anticipate the need to measure performance. Under time-to-market pressure. Starvation is the failure to schedule a component properly. the total time that passes for the computation. Similarly. How to Understand Performance Problems Learning to understand the performance of a running system is unavoidable for the same reason that learning debugging is.5. and many resources consumed. If you have a permanent log. There are many dimensions to the performance of a computer system. To do this you can use a profiling tool or a good log to figure out where the time or other resources are really being spent. However. Sometimes something that takes a little longer but doesn't burn up so many processor seconds will be much better in computing environment you actually have to deal with. or places where most of the resources are consumed. printlining can now be done in terms of the log records.6. you must build a mental model of why it is slow. Even if this contention does not occur.
and. Here are some suggestions: • • • • • • • Remove floating point operations. leaving in debugging statements that are no longer needed. or recursive functions. spend a few minutes considering if there is a way to remove it entirely. In the end. and attack that to get another two-fold improvement. Other examples include doing unnecessary I/O in inner loops. There is usually a way to do this. Fold constants together. you can reach higher. I've made errors such as failing to provide a relational database system with a proper index on a column I look up a lot. This is simple. This kind of improvement is sometimes called low-hanging fruit. before you argue for the redesign of a subsystem. Move I/O into a buffer. you should ask yourself whether or not your proposal will make it five to ten time better. inexpert use of libraries and other subsystems that are often poorly documented with respect to performance. Would a different algorithm do? Could you compute that while computing something else? If you can't find away around it.7. (This is a great opportunity to use your skills as a good programmer. instead of counting heads. Don't allocate new memory blocks unnecessarily. this will require not only ingenuity but also an understanding of the expense of each kind of statement and expression. After you've made a two-fold improvement in something. move stuff out. Move a pointer rather than recomputing indices. Often. For example. You can continue making small improvements or you can seriously redesign a system or a subsystem. Clear. How to Optimize Loops Sometimes you'll encounter loops. which probably made it at least 20 times slower. that take a long time to execute and are bottlenecks in your product. Consider the test and quality assurance effort that your change will require. 9 . meaning that it can be easily picked to provide some benefit.) However. 1. so it is much better to have a few big changes. On some systems compilers and hardware do these things for you. efficient code is better than code that requires an understanding of a particular platform. Try not to divide. unnecessary memory allocation. then you can optimize the loop. you need to at least rethink and perhaps reanalyze to discover the next-most-expensive bottleneck in the system. or chop the tree down. What do you do when you start to run out of low-hanging fruit? Well. the bottlenecks in performance will be an example of counting cows by counting legs and dividing by four. not only in the new design but also in convincing your boss that this is a good idea. Try not to do expensive typecasts. Each change brings a test burden with it.Beginner system or a significant part of it at least twice as fast. in particular. Before you try to make the loop a little faster. The cost of each of these operations depends on your specific system.
Caching brings with it the danger that the copy is sometimes can't reflect changes to the master instantaneously. if you are reading some data from a database and computing something simple from it. database queries. This is often in tension with other demands.Beginner 1. Memory allocated directly in the local variables of a subroutine. and. If you have long running systems memory may never be exhausted in testing but will be exhausted by the user. Caching is avoiding I/O (generally avoiding the reading of some abstract value) by storing a copy of that value locally so no I/O is performed to get the value. Techniques for doing this include using a binary representation instead of one that is human readable. A classic mistake is to use a hash table as a cache and forget to remove the references in the hash table. processors are fast compared to the cost of communicating with a hardware device. There is only one master---period. disk I/O. and can include network cost. You can ignore it for a while but eventually you will have to decide how to manage memory. For instance. you may have to explicitly deallocate memory yourself when it is about to become garbage. The creation of new objects is moderately expensive on any system. There are two very fundamental techniques to improving I/O: caching and representation. But even with garbage collection. 1. How to Manage Memory Memory is a precious resource that you can't afford to run out of. This is highly dependent on the kind of system you're working with. you can fill up all memory with garbage. Garbage collection is wonderful: it lessens errors and increases code brevity and concision cheaply. Since the reference remains. A third technique that is sometimes possible is to improve the locality of reference by pushing the computation closer to the data. Use it when you can. This cost is usually abbreviated I/O. transmitting a dictionary of symbols along with the data so that long symbols don't have to be encoded. More often you may be able to use a system that provides a garbage collector. A chunk of memory is useless. but you should explore it. however. things like Huffman encoding. at the extreme. You should look for and fix memory leaks early. Therefore building a fast system is often more a question of improving I/O than improving the code in some tight loop. Representations can often be improved by a factor of two or three from their first implementation. You should avoid unnecessary object creation. Depending on the system you use. the referent is noncollectable but useless. How to Deal with I/O Expense For a lot of problems. when nothing refers to it. This is called a memory leak.9. or even improving an algorithm. try to get the database server to do it for you. is usually cheap because the policy for freeing it can be very simple. 10 .8. and other use of some hardware not very close to the processor. Space that needs to persist beyond the scope of a single subroutine is often called heap allocated. such as a summation. The first key to caching is to make it crystal clear which data is the master and which are copies. A garbage collector notices garbage and frees its space without any action required by the programmer. file I/O. Representation is the approach of making I/O cheaper by representing data more efficiently. like human readability and portability. hence garbage.
try to reproduce the bug in a controlled way. We have multithreaded application servers in Java™ behind Apache™ web servers. I simply forgot to synchronize the garbage collector. before anything noticeable went wrong. Then. Resign yourself to that if the bug only occurs in production and not at your whim.Beginner An important case occurs when you can define an upper bound on the number of objects you will need at one time. good utilization of all the CPUs available (eight. The objects you need can be allocated and released inside this buffer in a set rotation pattern. set a trap for it by building a logging system.’ If that is not the source of variability. Every once in a while these would apparently get ‘stuck’ and cease doing anything useful. Then the queues emptied by these threads would quickly fill up all available memory and crash our server. so far as our logging allowed us to tell.10. so that you can guess at what the variability really is. try. Although after eight hours you will start to doubt it. The method may differ for each kind of object you create. or a buffer. The system could run a long time. Then you must apply careful intelligence to each chunk of allocated memory and design a way for it to be deallocated at the appropriate time. Try to record the circumstances under which the bug does occur. you have to wait for the bug to reoccur to get more information. you may be able to allocate a single block of memory. for hours. or even what the threads where doing when they got ‘stuck’. This cycle can go on for some time. a special one if you have to. At work we recently had an intermittent bug that took us several weeks to find. that can log what you guess you need when it really does occur. This is usually faster than heap allocation. to hold them all. I'm ashamed to admit I had begun to question the hardware before my mistake dawned on me. You can't debug it because you can't find it. in this case). such as reference counting. Try. This is so difficult that programmers often simply implement a rudimentary form or garbage collection. often finishing whatever task I began. to do this for them. 1. It took us about a week to figure this much out. this is may be a long process. the intermittent bug has to obey the same laws of logic everything else does. this was not in itself a giant problem---unless all four got stuck. I had very carefully insured correct concurrent evaluation of the functional program. when it would happen. yet often enough that it can't be ignored. the next suspect should be improperly synchronized concurrency. and we still didn't know what caused it. If you can't reproduce it. The hints that you get from the log may not provide the solution but may give you enough information to improve the logging. To maintain fast page turns. Since we had four threads. How to Deal with Intermittent Bugs The intermittent bug is a cousin of the 50-foot-invisible-scorpion-from-outer-space kind of bug. The improved logging system may take a long time to be put into production. This nightmare occurs so rarely that it is hard to observe. If these objects all take up the same amount of memory. 11 . You must make sure that every execution of a memory allocating operation is matched by a memory deallocating operation eventually. What makes it hard is that it occurs only under unknown conditions. so it is sometimes called a ring buffer. The stupidest intermittent bug I ever created was in a multi-threaded implementation of a functional programming language for a class project. such as ‘This only happens when we enter Wyoming as a value. we do all I/O in small set of four separate threads that are different than the page-turning threads. The condition may be related to data values. Sometimes you have to explicitly free allocated space so it can be reallocated rather than rely on garbage collection.
is like referring to surgery as ‘knife science’ --. but remember that design is an art. but not all. The kinds of experiments you will have to perform include: • Testing systems with small examples to verify that they conform to the documentation or to understand their response when there is no documentation. They move on to larger projects. 12 . Had these texts occurred commonly. It is natural and helpful to develop your own style.the harm was done: the topic became known as ‘computer science’---which. As he puts it referring to the 1960s[Knife]. How to Learn Design Skills To learn how to design software. How to Conduct Experiments The late. it took us weeks to finally understand and resolve the problem. 1. you can read some books on the latest design techniques.’ Although we had the source code (thank goodness!) we had not studied it carefully until by turning up the logging on our servers we finally realized that the email threads were getting stuck in the French stripper. Then you must do it yourself. the code was quadratic or worse. physicists do. Don't become dogmatic about particular design styles. 1. When you are finally done. study the action of a mentor by being physically present when they are designing.12. . just as some. The stripper performed well except on some long and unusual kinds of texts. We must work in the realm of experimentation. Design is a matter of judgment that takes years to acquire. This means that the processing time was proportional to the square of the length of the text. but most working programmers do not have the luxury of engaging in what Dijkstra means by computing science. great Edsger Dijkstra has eloquently explained that Computer Science is not an experimental science[ExpCS] and doesn't depend on electronic computers. People who write books on the subject have a vested interest in making it seem scientific. As it happens. it will be a great accomplishment of Computer Science. we would have found the bug right away. A smart programmer can learn the basics adequately in two months and can improve from there.and it was firmly implanted in people's minds that computing science is about machines and their peripheral equipment. Then study well-written pieces of software. we would never have had a problem.Beginner This illustrates some risk associated with third-party software.. Start with a small project. not a science. actually. hopefully in conjunction with other people.. We were using a licensed piece of code that removed HTML tags from text.11. On these texts. Due to its place of origin we affectionately referred to this as ‘the French stripper. consider how the design failed or succeeded and how you diverged from your original conception. If they had never occurred at all. After that. If thirty years from now programming can be performed without experimentation. Programming ought not to be an experimental science.
you should design your experiments to provide the most information possible. it is impossible to plan these effectively. and Collecting statistics that may hint at the solution to difficult or hard-to-repeat bugs. I can offer two bits of advice. it is about 50% likely that we will be done in five weeks (if no one bothers us during that time). Restate your assumptions. Managers love it. I don't think in this essay I can explain the design of experiments. This common interpretation problem requires that you explicitly discuss what the estimate means with your boss or customer as if they were a simpleton. and we must face up to it honestly. to predict accurately how long it will take to develop software is often lost on managers. especially if you find yourself confused or are working with others. really means: I promise to have it all done five weeks from now.Beginner • • • • Testing small code changes to see if they actually fix a bug. as people have a startling tendency to think wishfully that the sentence: I estimate that. Without predictability of the development time. However. no matter how obvious they seem to you. or the assertion that you are trying to test. Unfortunately. Therefore. You will often find yourself having to design a series of experiments. We are asked to do this impossible thing all the time. How to Estimate Programming Time 13 . but also planning the documentation. 2. However. it would be dishonest not to admit the impossibility of this task. this is in tension with keeping each experiment simple---you will have to develop this judgment through experience.2. and when necessary. Checking the integrity of data. try to be very clear about your hypothesis. if I really understand the problem. each of which is based on the knowledge gained from the last experiment. both theoretically and practically. marketing. as well they should. you will have to study and practice. Team Skills 2. In a commercial project it also requires sales and finance. The fact that it is impossible.1. deployment. There is a lot of room for miscommunication about estimates. explain it. Why Estimation is Important To get a working software system in active use as quickly as possible requires not only planning the development. It also helps to write the hypothesis down. First. 2. Good estimation provides predictability. Measuring the performance of a system under two different conditions due to imperfect knowledge of there performance characteristics.
People vary widely in skill. It's far better to give proper visibility into what you are actually doing. It takes so much labor it may be a good idea to estimate the time it will take to make the estimate. Unk-unks by definition cannot be estimated individually. For instance. There are of course. and the appearance of slacking or overestimating is born. you should try to have the people who will do the work do the estimate. And what is the probability of a mandatory company-wide training seminar? If it can be estimated. it is your duty to state so forcefully enough that your manager does not commit to them and then become embarrassed when the risk occurs. For instance. One person is unlikely to be sick in any given week. It also takes labor. I know good engineers who pad estimates implicitly. and stalling certainly will displease the stalled. Calamity strikes when a strong programmer estimates for herself and then weak programmers are held to this estimate. at least do an average weighted by your estimates of the probabilities.Beginner Estimation takes practice. Most engineers are enthusiastic and eager to please. While stalling. If there are big risks that cannot be evaluated. but I recommend that you do not. If a task will probably take one day---but might take ten days if your approach doesn't work---note this somehow in the estimate if you can. or handle them in some other way that you communicate to your boss. unknown unknowns. Restate that meaning as the first and last part of your written estimate. or two days working on some other useful project. or unk-unks. and confidence. and you should try to have team-wide consensus on estimates. let your boss forget that they exist. as well as allowing the opportunity for tactical reassignment of resources (for instance. shifting burden away from weaker team members to stronger). an engineer might estimate three days for a task that she truly thinks will take one day. The engineer may plan to spend two days documenting it. When asked to provide an estimate of something big. In a team environment. especially if you are asked to estimate something big. documentation. When not possible to take the time for some investigation. Any risk factor that you can identify and assign an estimate to should go into the schedule. One of the results of padding is trust in you may be depleted. tremendous advantage is gained by making this visible to the manager. Hopefully in such a case whatever is needed will be done to decrease the risk. You can try to create a global line item for all unk-unks. and it makes real progress. time for planning. But a large project with many engineers will have some sick time. however. 14 . If political pressure permits. experience. preparedness. But it will be detectable that the task was done in only one day (if it turns out that way). you should first establish the meaning of the estimate very clearly. likewise vacation time. and vacation time are all very important. this is the most accurate way of producing the estimate. If documentation takes twice as long as coding and the estimate says so. Prepare a written estimate by deconstructing the task into progressively smaller subtasks until each small task is no more than a day. if not. ideally at most in length. The most important thing is not to leave anything out. This gives your boss visibility into what is using up your time at a minimum. it may be possible to consider doing or prototyping the task. The act of having the whole team agree on a line-by-line basis to the estimate clarifies the team understanding. put a line item for that in the estimate. Pad explicitly instead. time for communicating with other groups. If you spend part of each day dealing with knuckleheads. testing. stick it in. But an on-the-spot estimate probably won't be accurate and honest. and it is devilishly easy for an estimate to become a schedule without the unk-unks considered. and might get you more time. You cannot. the most honest thing to do is to stall.
go to the library (the physical building in which books are stored). For example. you should talk to a LISP expert and a database expert. you must still search the internet and the library. How to Find Out Information The nature of what you need to know determines how you should find it.Beginner If you can convince your company to use Extreme Programming. You can even learn important things. from the Internet. If you need general knowledge about something subjective the history of what people have thought about it. and this is both more fun and more productive. 2. 2. talk to someone working in that field. If you want to make a personal decision that only you can make like whether or not you should start a business. If you need information about concrete things that are objective and easy to verify.3. both by enjoying your presence and hearing the particular question. ask a large number of people politely by searching the internet for it or by posting on a discussion group. Suppose you have studied the idea from all angles. If you need information that no one else could be expected to know for example. as they each present a complete ambiguous and random pattern that your own subconscious will assign meaning to. ‘does this software that is brand new work on gigantic data sets?’. The person learns about you. How to Utilize People as Information Sources Respect every person's time and balance it against your own. If you want an opinion or a value judgment that takes into account some unique circumstance. and you may learn the answer you seek. go to the library. if you want to know whether or not it is a good idea to build a modern database management system in LISP. If you want to know how likely it is that a faster algorithm for a particular application exists that has not yet been published. and yet still remain indecisive.4. but you can easily spend more time searching and sorting the results and attempting to divine the authority of the results than it would take to read the pertinent part of a solid book. The multitude of available divination techniques are very useful for determining your own semi-conscious desires. like install a software package. for example the latest patch level of a software product. talk to an expert. have done all your homework. This is usually far more important than your question. you will only have to estimate relatively small things. and worked out all the consequences and pros and cons in your mind. For instance. You might learn how to do something trivial. If that fails. If you need to know how to do something that is not trivial get two or three books on the subject and read them. like good programming technique. After those options are completely exhausted. consider divination. to learn about math or mushrooms or mysticism. you may design an experiment to ascertain it. try putting into writing a list of arguments for and against the idea. You now must follow your heart and tell your brain to shut up. Asking someone a question accomplishes far more than just receiving the answer. Don't search on the internet for anything that smacks of either opinion or subjective interpretation: the ratio of drivel to truth is too high. 15 . You learn about the person in the same way.
Beginner However. the less benefit they derive. but you should talk to your boss's boss a little. A good programmer does not often need the advice of a Vice President of Sales. It may be uncomfortable. I once asked to listen in on a few sales calls to better understand the job of our sales staff. study them. what they need to get out of it. 2. the value of this diminishes the more you do it. the Golden Rule is all you really need: ``Do unto others as you would have them do unto you. nobody will read it. say ``yes'' and quietly begin looking for a better job. But even if you are a lousy writer or have poor command of the language in which you must document. I strongly believe that an executive of 100 people should spend five minutes a month talking to each person in her organization. keeping the benefit in balance with the time spent. good writing. which would be about 5% of their time. by that factor. A summer intern in a highly technical position can't be expected to accomplish too much. So why is this tolerated? Because the pestered are receiving something important from the intern. and five minutes is too much if they have one thousand employees. but even if this is not the case there is much to gain. Managers often don't understand this. like testing. but I believe you have a duty to talk a little bit to all your superiors. It is important to respect your own time. you be sure to ask for it. and to get the benefit of communicating with them. maybe. Writing good documentation is. to the tribe. I suggest you find books on writing. can take many times longer than developing code. will save you a great deal of time. if you write crap. The benefits of communication must be weighed against the costs. but if you ever do. A strange example of this is the summer intern. It is your job to provide them this benefit. even if it will cost them time. and the more they talk to you. first of all. they can be expected to pester the hell out of everybody there. and practice. But ten minutes might be too much. There's nothing quite as effective as putting an accurate estimate of the amount of time it will take to produce good documentation into an estimate to slacken the demand for documentation. using the most precious commodity a person has: their time. each month. They may also be trying to recruit the intern. How to Document Wisely Life is too short to write crap nobody will read. they get a chance to see things from a different perspective. after all. If talking to someone. They get a chance to hear some new ideas. the particular costs and benefits derived differ from person to person. The basic rule is that everyone benefits from talking to you a little bit. then you should do it unless you think their time is more valuable than yours. Therefore a little good documentation is best. They get a chance to showoff a little. no matter what. This flatters them and you will learn something and teach them something. If someone absolutely insists that you write truly useless documentation. You should ask people for a little bit of their wisdom and judgment whenever you honestly believe they have something to say. If you do that. This took no more than 30 minutes but I think that small effort made an impression on the sales force. The truth is cold and hard: documentation. You are. you 16 . because even bad documentation gives them a false sense of security that they are not dependent on their programmers. The amount of time you spend talking to each person in your organization depends on their role (more than their position).'' Take time to really think about who will be reading your documentation. Furthermore. You should talk to your boss more than your boss's boss.5. and how you can teach that to them.
To understand it takes learning time. and a good programmer. The source code can at worst be wrong and confusing. It is important to remember that abstraction and encapsulation. if not written perfectly. two of a programmer's best tools. and that time will have to come out of some schedule. this seems easier to the experienced programmer than to the beginner. anyone who needs to see code-level documentation will in most cases be able to and prefer to read the code anyway. How does one write self-explanatory code? What does that even mean? It means: • • • • • • Writing code knowing that someone will have to read it. You will probably have to experiment with it. Would it actually save time to rewrite some of it? Could you trust it better if you rewrote it? Be careful of arrogance here. the quality of that code should affect your perception of the risk of problems and unk-unks. consider what it would take to rewrite some or all of the code. but will it really be easier for the next person who has to read it? If you rewrite it. Choosing a solution that is straightforward. because the act of trying to document the code will force you to consider angles you might not have considered. To understand it. in order to work with unclear code you must understand it. even if you could get by with another solution faster. More importantly however. or ``doIt''! 2. When it comes to actually documenting code itself. There are two good reasons for this. until you have walked in their shoes. Applying the golden rule. This is a good time to document. you will have to read the source code. Admittedly. and you must insist on it. Thinking about the reader and spending some of your precious time to make it easier on her. what will the test burden be? Will the need to re-test it outweigh any benefits that might be gained? In any estimate that you make for work against code you didn't write. How to Work with Poor Code It is very common to have to work with poor quality code that someone else has written. somewhere. Don't think too poorly of them. If you rewrite it. This does not make it easier on the responsible programmer. Regardless. even if it is only for yourself. however.Beginner will be an above average documentation writer. 17 . The documentation. and Not ever using a function name like ``foo''. it will be easier for you to deal with.6. is that the code and the documentation cannot be inconsistent if there is no documentation. Sacrificing small optimizations that obfuscate the code.``bar''. as opposed to producing documents that can actually be read by non-programmers. can lie. the best programmers I've ever known hold a universal sentiment: write self-explanatory code and only document code in the places that you cannot make it clear by writing the code itself. While you're doing this. First. and that is a thousand times worse. and the resulting document may be useful. They may have been asked very consciously to get something done quickly to meet schedule pressure.
Committing a mistake that slows down your teammates is a serious error. I sometimes meditate for 15 minutes when stumped and the problem magically unravels when I come back to it. 2.8. or in a branch of its own. 2.9. the testing of an individual piece of coded functionality by the team that wrote it. I was late to appreciate the benefits of source code control systems but now I wouldn't live without one even on a one-person project.10. you can try to wall off the parts that are particularly bad so that they may be redesigned independently. I can do no better than to recommend their writings. In particular. official code that will be shared with the team or released. It's possible that temporarily switching to any other activity may work. How to Unit Test Unit testing. Part of designing the code is designing how it will be tested. take a break. You should write down a test plan. How to Use Source Code Control Source code control systems let you manage projects effectively. Sometimes the test will be simple: ``Does the button look good?'' Sometimes it will be complex: ``Did this matching algorithm return precisely the correct matches?'' Use assertion checking and test drivers whenever possible. A good technique for using a source code control system is to stay within a few days of being upto-date at all time.Beginner are particularly applicable to lousy code. Since each change is marked as a new revision with a new name or number. Take Breaks when Stumped When stumped. Generally they are necessary when you have team working on the same code base. but in a way that it is inactive and will not be called. organic system. 2. 2. even if it is only one sentence. They track all changes in different versions so that no code is ever lost and meaning can be assigned to changes. and therefore not create any problems for anybody else. A night's sleep sometimes does the same thing on a larger scale. The Extreme Programming developers are writing extensively on unit testing effectively. is a part of coding. However. They're very useful for one person and essential for a group. This not only catches bugs early. it is often taboo. Code that can't be finished in a few days is checked in. but if you can add a certain amount of abstraction to it you can obtain some of the benefits of a good design without reworking the whole mess. one begins to think of the software as a visibly progressive series of improvements. since the code you modify is kept carefully separate from committed. they have another great advantage: they encourage thinking about the code as a growing. not something different from it. but is very useful later on and lets you eliminate mysteries that you would otherwise have to worry about. The unfortunate fact is that it is not a 18 . One can create throw-away and debugging code with confidence with a source code control system. I think this is especially useful for beginners. How to Recognize When to Go Home Computer programming is an activity that is also a culture.7. You may not be able to redesign a large block of code.
it is not a programmer's duty to be a patsy. If you are tempted to be dishonest or deceptive in a way that you normally are not due to fatigue. that it is stupid to work so much that you are getting little out of that extra hour you work. There can't be any fixed rules for solving this problem. I don't know if it is fair to expect 60 hours of work from a person. Since I have children. and Quit if you have to. such as reserving certain days in which you will attend only the most critical meetings. but it is the best compromise I have been able to work out. Most programmers are good programmers. However. and sometimes when to suggest that other people go home. who is responsible not only for themselves but their teammates as well. anymore than there can be fixed rules for raising a child. This means that often much more than that is required. Each programmer needs to do whatever it takes to procure efficient work periods. Don't use cocaine or amphetamines to combat fatigue. To do that. You should go home if you are thinking suicidal thoughts. There is a certain amount of mental inertia associated with getting warmed-up to a problem and deeply involved in it. and that is sometimes expected of me. and 50 is pretty much a minimum. However. This is serious problem for a good programmer. I don't think you can trust all the stories you hear. How to Deal with Difficult People 19 . they have to manage their time effectively. For me personally. but I think 60 hours a week is common. and say no as a team when necessary. computer programmers are traditionally overworked. however. you should take a break. that's any more than 60 hours a week. for example a manager trying to impress an executive. The sad fact is programmers are often asked to be patsies in order to put on a show for somebody.11. I am sure. There are four defenses against this: • • • • Communicate as much as possible with everyone in the company so that no one can mislead the executives about what is going on. You have to recognize when to go home. Go home if you have a contagious disease. For both cultural/historical reasons (the need to work at night on unloaded computers. Programmers often succumb to this because they are eager to please and not very good at saying no. I don't even know if 40 is fair. people must sleep and perform other duties. uninterrupted blocks of time in which to get warmed-up and concentrate. Learn to estimate and schedule defensively and explicitly and give everyone visibility into what the schedule is and where it stands. Beyond 60 hours a week is an extraordinary effort for me. You should take a break or go home if you think homicidal thoughts for more than a few seconds. Many programmers find they work best when they have long. for the same reason---every human being is different. Learn to say no. Don't abuse caffeine. and good programmers want to get a lot done. I personally think a programmer should exercise noblesse oblige and shoulder a heavy burden. I try to spend evenings with them sometimes. Each person needs to find a way to satisfy both their human rhythm and their work rhythm. for example) and because of overwhelming timeto-market pressure and the scarcity of programmers. which I can apply for short periods of time (about one week). The rhythm that works best for me is to work a very long day. sleep in the office or near the office (I have a long commute from home to work) then go home early enough the next day to spend time with my children before they go to bed. 2.Beginner culture that values mental or physical health very much. You should send someone home if they show serious mental malfunctioning or signs of mental illness beyond mild depression. I am not comfortable with this.
Difficult people do change and improve. You may even be a difficult person yourself. It is critical that you listen and understand the difficult person without prejudice caused by the person. Don't let a bully force you to do something you don't agree with. They are more prone to duck work and resist passively than others. do it the other way whole-heartedly. One of the challenges that every programmer but especially leaders face is keeping the difficult person fully engaged. After a reasonable period of trying to understand. If you are the kind of person who has a lot of conflicts with coworkers and authority figures. but it is very rare. 20 . The key is to respect them appropriately. This can be very disturbing to some programmers who have no experience in this sort of thing and whose previous life experience has taught them patterns of behavior that are not useful in the workplace. Try to keep this communication cool and cordial. you should cherish the independence this implies. do what you think is best. make a decision.Beginner You will probably have to deal with difficult people. Don't make a decision for any personal reasons. but work on your interpersonal skills without sacrificing your intelligence or principles. I've seen it with my own eyes. and be prepared to explain the reasons for your decision. Difficult people are often extremely intelligent and have something very useful to say. If you are the leader. If you are a teammate with a difficult person. Programmers have to work together as a team. When disagreement arises. it cannot be ducked for long. everyone has transitory ups and downs. If it doesn't go your way. However. don't let the leader's decision have any personal impact. which is more than you will want to but not as much as they might want. and don't accept any baits for greater conflict that may be offered. A failure to communicate is often the basis of disagreement but it can sometimes be removed with great patience. it must be resolved somehow. Difficult people are often inured to disagreement and they are less affected by social pressure to compromise than others.
For example. Intermediate 1. in each project. 1. is personally motivating to me. they will have low morale. however. If programmers are asked to do something that is not beautiful. You establish trust by being responsive and informative to those outside your department or team. With people that are not teammates. How to Tradeoff Time vs. and ask for unreasonable favors. Don't pretend to know something that you don't. You don't need to know 21 . however small. fun will make the most money for the company. This has practical and important consequences. Look for opportunities to apply new techniques. you may have to make a clear distinction between ``not knowing right off the top of my head'' and ``not being able to figure it out. The things that are specific to programming that I can identify are: • • • Use the best language for the job. this should not be an issue. How to Stay Motivated It is a wonderful and surprising fact that programmers are highly motivated by the desire to create artifacts that are beautiful. stupid. or nifty. Occasionally someone will abuse this trust. Relating each bug to a happy customer.'' 1. languages. ever. and boring stuff. measure the impact of your work in terms of something that will be personally motivating. and technologies. there are entire industries organized around motivational techniques some of which apply here. Don't be afraid of this.Chapter 3. Personal Skills 1. if possible. but in the end.3. useful. How to be Widely Trusted To be trusted you must be trustworthy. but you can't be a good intermediate programmer without knowing basic computational complexity theory. Obviously. because it is independent of the number that may still exist. no trust will be invested in you. Space You can be a good programmer without going to college.1. This desire is not unique to programmers nor universal but it is so strong and common among programmers that it separates them from others in other roles. such as your teammates. Finally. With those close to you. or nifty. If know one knows about you. You must also be visible. and is also affects the total value I'm adding to my company's customers in only the smallest possible way. when fixing bugs. just explain what you would have to give up doing to perform the favor. There's a lot of money to be made doing ugly. Try to either learn or teach something. counting the number of bugs that I have fixed is not at all motivational to me.2. useful.
but there is almost always a ‘wall’. but one can at least create 500 simulations and try to model some part of what they might do. that is. 22 . There may be some exceptions. The purpose of stress testing is to figure out where the wall is. it is common that the system does work under a load but fails to work in some way when the load is heavy enough. Improving the space/time tradeoff can often change one or the other dramatically. at the expense of more computation time when you have to decode them. but I personally think you should be able to understand the difference between ``constant-time''. To me. derived from ‘to hit’. computational complexity theory is beautiful and as profound as physics---and a little bit goes a long way! Time (processor cycles) and space (memory) can be traded off against each other. but then failure is catastrophic. but you can't let that blind you to the cold hard fact that improving something that is not a problem will not make any noticeable difference and will create a test burden. rather than gradual. You might be able to intuit how to tradeoff time against space without this knowledge. At first it appears that the purpose of stress testing is to find out if the system works under a load. Memory on modern computers appears cheap. of course. You can save time by caching. Consider this carefully before you trade away space to gain speed. degradation of performance or spirit. because it often helps to clarify exactly what is expected. and then figure out how to move the wall further out. A plan for stress testing should be developed early in the project. and this is a fine example. but one must know the answer when designing the system that answers the request. Engineering is about compromise. It isn't really possible to simulate 500 erratic and unpredictable humans using a system concurrently very easily. depends. It is not always systematic. In reality. throughout some of the British Commonwealth. however. However. the amount of time it takes to run is sometimes a function of the size of the input. I call this hitting the wall or bonking1. 1This term has several meanings. It's fun to work on an algorithm. we can say an algorithm's worst/expected/best-case running time is ``n log n'' if it is proportional to the size (represented by the variable n) times the logarithm of the size. before you work on this you should ask yourself if what you are improving is really the thing that needs the most improvement. The notation and way of speaking can be also be applied to the space taken up by a data structure.4. When that is true. This usually cost a small amount of space but may complicate the algorithm. In designing or understanding an algorithm. Is two seconds for a web page request a miserable failure or a smashing success? Is 500 concurrent users enough? That. I have been told it has an additional. because unlike processor time. 1. one can save space by encoding things more tightly. The stress test needs to model reality well enough to be useful. You can sometimes save time by maintaining more information in a data structure. There are also other hidden costs to using memory. you can't see it being used until you hit the wall. but is particularly used by athletes to describe running out of blood sugar or some other basic resource that manifests as a sudden. but in its absence you will not have a firm basis for communicating with your colleagues.``n log n'' and ``n squared''. In general. and the time to allocate and deallocate it. spending space to store a local copy of something. and very different meaning.Intermediate ``big O'' notation. at the expense of having to maintain the consistency of the cache. How to Stress Test Stress testing is fun. such as your effect on other programs that must be resident.
it is an accepted style to hide an integer variable on an object behind mutators and accessors. the code itself becomes messy. only the little interface to it. so that the variable itself is not exposed. Input/Output. increasing the maximum load the system can handle. If the wall is too close to satisfy your needs. figure out which resource is the bottleneck (there is usually a dominant one. start out with a light load and load the system along some dimension---such as input rate or input size---until you hit the wall. however. For example. It is relatively easy and certainly a good idea to confine non-portable code to designated areas. or data contention? Then figure out how you can move the wall. But I don't think the benefit of this outweighs the cost of the wordiness of it when my team owns the calling code and hence can recode the caller as easily as the called. such as a class that makes database queries that are specific to a given DBMS. Usually performance under heavy load is more important than performance under a light load. and is perhaps appropriate to a library writer who must publish a very stable API. Visibility into these systems. or simply easily ported? I think a non-portable. This does allow the implementation of that variable to be changed without affecting the calling code. shortand-easily-ported piece of code is better than a long portable one. logging often gives a good idea of the wall-clock time between two events in the system. processor. For instance. no single technique is sufficient. might not help or might actually hurt the performance of a lightly loaded system. a number of computers and many software systems may be cooperating. that is. I personally think. 1. This is a form of speculative programming. the performance is non-linear in the size of the input) these other software systems may be a bottleneck. can be very helpful. Should code be portable to a different computer. The attraction of this is understandable but the value of code brevity must be measured against the value of abstraction. 23 . Portability poses a similar problem. that you should not produce much speculative code. network bandwidth. software system or platform. As the project progresses and fatigue sets in. in a modern system. I strongly recommend the article ``Succinctness is Power'' by Paul Graham[PGSite].5. Four or five extra lines of code is a heavy price to pay for this speculative benefit. Note that moving the wall. You may have to get visibility into several different dimensions to build up a mental model of it. doesn't give visibility into memory utilization or even data structure size. but also to providing predictability so that the business can be managed effectively. one sees a mistake made by enthusiastic idealists: at the start of the project a lot of classes are defined that seem wonderfully abstract and one may speculate that they will handle every eventuality that may arise. but unless carefully constructed. The empty classes are a burden to document that is ignored when under pressure. Particularly when you are hitting the wall (that is. There is a certain dogma associated with useful techniques such as information hiding and object oriented programming that are sometimes taken too far. even if only measuring the processor load on all participating machines. Function bodies become longer than they should be. compiler. The final result would have been better if the energy spent on abstraction had been spent on keeping things short and simple. Beginning programmers in their enthusiasm often create more abstraction than is really useful.Intermediate In stress testing. Similarly.) Is it memory. One sign of this is if you create classes that don't really contain any code and don't really do anything except serve to abstract something. You should carefully choose how abstract you need to be. How to Balance Brevity and Abstraction Abstraction is key to programming. Knowing where the wall is is essential not only to moving the wall. Occasionally. These techniques let one code abstractly and anticipate change.
How to Do Integration Testing Integration testing is the testing of the integration of various components that have been unit tested. no matter how good you are. Communication Languages There are some languages. When learning a new programming language.9. How to Learn New Skills Learning new skills. try to manage a small one first. You must include time for this in your estimates and your schedule.6. but understand that it often not much better than the same amount of time spent simply playing with the new skill you want to learn. Learn to Type Learn to touch-type. however. What can you offer a potential mentor in exchange for their knowledge? At a minimum.8. 1. especially non-technical ones. This is an intermediate skill because writing code is so hard that the speed at which you can type is irrelevant and can't put much of a dent in the time it takes to write code. that is. 24 . and communication. When learning to manage a software project. Legend has it that when Michael Tiemann2 was at MCC people would stand outside his door to listen to the hum generated by his keystrokes which were so rapid as to be indistinguishable. it takes dedicated time that is not much fun to learn something like that. 1. Book-reading and class-taking are useful. A good mentor is no replacement for doing things yourself. you should offer to study hard so their time won't be wasted. is the greatest fun of all. Most companies would have better morale if they understood how much this motivates programmers. Try to get your boss to let you have formal training. If it is unavoidable estimate it carefully.Intermediate 1. It is far better to gradually integrate things as they are completed over the course of the project. However. It is. you have to put yourself in a forgiving position where you can exercise that skill. honesty. Give your people a chance to play and practice courage. formally defined syntactic systems. But could you have any respect for a programmer who had never written a program? To learn any skill. understand how they learn and assist them by assigning them projects that are the right size and that exercise skills they are interested in. Integration is expensive and it comes out in the testing. Humans learn by doing. Don't forget that the most important skills for a programmer are not the technical ones. that are not programming languages but communication languages---they are designed specifically to facillitate communica2At the time of this writing Michael Tiemann is the CTO of RedHat. easier to ask for training than playtime in our imperfect world. If you lead people. even though a lot of formal training is just sleeping through lectures waiting for the dinner party. 1. Ideally you should organize a project so that there is not a phase at the end where integration must explicitly take place. This is a fun test of your commitment. but is a lot better than a book. by the time you are an intermediate programmer you will probably spend a lot of time writing natural language to your colleagues and others. try to do a small project it in before you have to do a large project.7.
Make sure your plan includes time for: internal team meetings. dealing with outsiders. though you sometimes see it presented as if it was. structuring the representation into a linear sequence and parsing back into a structure. which are less important than the standardized core. scheduled periodic activities. How to Manage Development Time To manage development time. maintain a concise and up-to-date project plan. 2. A project plan is an estimate. The project plan exists to help make decisions.Intermediate tion through standardization. a schedule. vacations. and SQL. Team Skills 2. SQL is the lingua franca of relational databases. If the project plan is either too long or not up-to-date. the project plan should be a consensual agreement. UML is a rich formal system for making drawings that describe designs. It is not a solution to data interchange problems.and correctness-checking. If you miss a milestone. these decisions are about individual persons. this creates the illusion that you might be able to make up the days you missed in the latter part of the project. and I find myself using a simpler boxes and arrows style for design drawings. You need to know about it because designs are sometimes communicated in it. The project plan can serve as 25 . You may or may not work in any field that can benefit from an understanding of relational databases. but you should have a basic understanding of them and they syntax and meaning of SQL. maintenance of existing products. SQL is a very powerful and rich data query and manipulation language that is not quite a programming language. The estimate and schedule could never have been perfect to begin with. XML is a standard for defining new standards. though again only a fraction of what you are likely to need in practice. If you use a fancy project planning tool. But I'm fairly sure UML is at least as good for you as studying Latin. and an assignment of your team or your own time to each task on the estimate. documentation. demos. and maintenance of the development environment. namely. typically quite product-dependent. integration testing. but use it maintain concision and up-to-dateness. If you are on a team. it will be useless for making decisions. The plan and your judgment let you decide if you should shift tasks from one person to another. In 2003 the most important of these are UML. you should take immediate action such as informing your boss that the scheduled completion of that project has slipped by that amount. It should also include other things you have to remember to do. Rather. sickness. It has many variations. capable of conveying a great deal of information if both the author and the audience know UML. such as meeting with the quality assurance people. It provides some nice type. or ordering equipment.1. The milestones mark your progress. do not be seduced into creating a Big Design Up Front (BDUF) for the project. a set of milestones for marking progress. In a lot of cases UML is too formal. In reality. You might. Therefore the scheduled completion of the project has already slipped. both at the start and as you go. preparing documentation. It's beauty lines in that is both visual and formal. But it is just as likely that you have underestimated that part as that you have overestimated it. There are very helpful tools for making UML drawings that look very professional. it is a welcome automation of the most boring part of data interchange. whether you like it or not. not to show how organized you are. You should have some familiarity with all of these so that you can communicate well and decide when to use them. XML.
If you are considering using third-party software. they often know more about specific technologies and even programming techniques than you will. but don't rely on them. Never let it be written down in any documents used by your company. it is still risky. However. The best way to use them is as educators in-house that can teach by example. They are wonderful people and deserve a great deal of respect. you must review it carefully as you go along. but at least it is a risk that can be tackled. It is unwise to be merely skeptical of a software company's promise to release a certain product with a certain feature at a certain date. it is far wiser to ignore it completely and forget you ever heard it. Since they get to see a lot of different projects. really. but you will usually have more knowledge of the team members closer to you. This is true of all team members. They may have less to gain if the company does well. They can move more easily. If third-party software is not vapor. How to Communicate the Right Amount Carefully consider the cost of a meeting. and some will be bad. 2. Often times a project will depend on a third-party software system so completely that if the integration fails the project will fail. People might not like to hear that it will take two weeks or two months to evaluate each of three products for suitability. Never let a schedule depend on vapor. How to Manage Consultants Use consultants. Try to have a contingency plan. you should devote energy early on to evaluating it. You cannot get to the end of the a project with the risk of a large block of code that has not been reviewed. Express risks like that clearly in writing in the schedule. You can save a lot of time if you can find those experts. rest any hopes on vapor. if only because you may not have enough time to learn their strengths and weaknesses. Understanding the suitability of existing third party software for a particular purpose is very tribal knowledge. This is the surest way to go out of business. How to Manage Third-Party Software Risks A project often depends on software produced by organizations that it does not control. The cost of integrating cannot be accurately estimated without a proper evaluation. some will be average.2.3. Never. but hopefully your selection of consultants will not be as careful as your selection of employees. ever.Intermediate a way to give outsiders or your boss a view into what you or your team is doing. such as another system that can be used or the ability to write the functionality yourself if the risk can't be removed early. For this reason it should be short and up-to-date. Some will be good. If consultants are going to write code. so you will get more bad ones. it costs its duration multiplied by the number of partici- 26 . they usually cannot become part of the team in the same sense that regular employees are. Their financial commitment is much lower. It is very subjective and generally resides in experts. but it has to be done as early as possible.4. There are great risks associated with third party software that must be recognized by everyone involved. 2. 2. Vapor is any alleged software that has been promised but is not yet available.
say so.1. you must remember that you will never be able to say ‘I told you so!’ since the alternate decision was never fully explored. In that case there is nothing more to say. Usually. and you should decide whether you will stand behind the decision even though you disagree with it. If it is a large mistake in you opinion.5. How to Tradeoff Quality Against Development Time Software development is always a compromise between what the project does and getting the project done. 3. then of course you must present an argument. Whether the decision is reversed or not. You should then evaluate whether to raise the issue on the basis of the benefit to the company or tribe. If it is a small mistake in your opinion. 2. This shows how valuable you are because you are independent and are not a yes-man. Meetings are sometimes necessary. but it should be handled delicately. Make it clear what is being lost and what is being gained. Sometimes a decision that you disagree with will be made when the decision makers did not have the full benefit of you opinion. In the worst of circumstances the decision maker is insecure and takes it as a personal challenge to their authority. How to Disagree Honestly and Get Away with It Disagreement is a great opportunity to make a good decision. You should present your argument in private. this is not a problem. your understanding of it should be much better than your boss's understanding. After all. but respectful of the decision and a team player. The quality of communication in small meetings is better. For instance. and less time overall is wasted. but smaller is usually better. some very good programmers lack the confidence needed to challenge a decision even when they have good reason to believe it is wrong. More useful work is done during lunches with colleagues than during any other time. you may be asked to do something that is a poor software engineering practice and that will lead to a lot of maintenance problems. If this happens your first responsibility is to inform your team and to clearly explain the cost of the decrease in quality. It is best to remember that in such circumstances people react with the reptilian part of their brains. For example. It is a shame that more companies do not recognize nor support this fact. take this as a sign that the meeting should be smaller. Judgement 3. it may not be worth reconsidering. Everything possible should be done to encourage informal communication.Intermediate pants. and try to show how new knowledge changes the basis on which the decision was made. and at what cost the lost 27 . If you can support this decision even though you disagree. But you may be asked to tradeoff quality to speed the deployment of a project in a way that offends your engineering sensibilities or business sensibilities. In some stressful circumstances and with some personality types this can lead to things being taken personally. If any one person is bored at a meeting. Hopefully you feel that you have expressed your thoughts adequately and been heard before the decision is made.
whether created in house or by a third party. With source code. If the quality tradeoff affects the quality assurance effort. Explain this to your team so that they can plan for it. Having the source code for a component decreases the risk by a factor of four. you can evaluate it easier.Intermediate ground will be regained in the next cycle.2. it is very important to consider if it is really mature enough to be used. find workarounds easier. point that out (both to your boss and quality assurance people). 3. then the eventual rewrites will be far more painless. the visibility provided by a good project plan should be helpful. Here are ten questions you should ask yourself about it: 1. If she still insists you should try to isolate the shoddiness into particular components that you can plan to rewrite or improve in the next cycle. but it makes porting easier. If the component proves to be completely unworkable. point that out. debug it easier. into a usable product. If good interfaces and abstractions exist throughout the code. otherwise you will uncomfortably have to maintain an unofficial version . This increases productivity through synergy and reuse. Encapsulation is not portability. Is it vapor? (Promises are very immature). How to Manage Software System Dependence Modern software systems tend to depend on a large number of components that may not be directly under your control. If you make fixes. Before you consider integrating with a software system. NinjaProgrammer at Slashdot sent in this gem: Remember that a good design will be resillient against poor code implementations. One of the biggest risks is the period of bugginess and near inoperability that is often associated with software before it matures. each component brings with it some problems: • • • How will you fix bugs in the component? Does the component restrict you to particular hardware or software systems? What will you do if the component fails completely? It is always best to encapsulate the component in some way so that it is isolated and so that it can be swapped out. 28 . If it is hard to write clear code that is hard to fix. In this. which is almost as good. If the quality tradeoff will lead to more bugs being reported after the quality assurance period. you should give them to the owner of the component and get the fixes incorporated into an official release. How to Decide if Software is Too Immature Using software other people wrote is one of the most effective ways to quickly build a solid system. consider what is wrong with the core design that is causing this. and make fixes easier.3. It should not be discouraged. However. 3. but you may have to write your own. but the risks associated with it must be examined. you may be able to get a different one. through usage.
You should also 29 . but do not invest in a solution bigger than your own project without conscious thought. 9. 8. build decisions. This will force you to consider the integration costs. 3. How to Make a Buy vs.Intermediate 2. • • • • • • • How well do your needs match those for which it was designed? What portion of what you buy will you need? What is the cost of evaluating the integration? What is the cost of integration? What is the cost of evaluating the integration? Will buying increase or decrease long term maintenance costs? Will building it put you in a business position you don't want to be in? You should think twice before building something that is big enough to serve as the basis for an entire other business. you may wish to change your business plan. and engineering savvy. one for building and one for buying. Can you hire people to work on it even if it is bad? A little consideration of these criteria demonstrates the great value of well-established free software and open-source software in reducing risk to the entrepreneur. 6. This requires a great combination of business. Is there an accessible body of lore about the software? Are you the first user? Is there a strong incentive for continuation? Has it had a maintenance effort? Will it survive defection of the current maintainers? Is there a seasoned alternative at least half as good? Is it known to your tribe or company? Is it desirable to your tribe or company? 10. 3. After considering these questions. 5. 4. Build Decision An entrepreneurial company or project that is trying to accomplish something with software has to constantly make so-called buy vs.4. you should perhaps prepare two draft project plans. If their idea is compelling. Such ideas are often proposed by bright and optimistic people that will have a lot to contribute to your team. 7. management. This turn of phrase is unfortunate in two ways: it seems to ignore open-source and free software which is not necessarily bought. Even more importantly. it should perhaps be called an obtain and integrate vs. build here and integrate decision because the cost of integration must be considered.
If you want to become a team leader. This serves a very valuable purpose of possibly revealing character flaws and weaknesses under stress. like writing well. Evaluate yourself. Candidates are no more honest with interviewers than they are with themselves. If there are several buy decisions under consideration. If you want to become a manager. Good people want to be hired for their skills. If that is too much to try. Express appreciation for people's contribution to the success of the larger organization. Some are torturous. but in the heat of the interview it is hard to recognize. To estimate the integration costs. some energy will have to be spent evaluating each. since this frees them up to take greater responsibility. If you can't evaluate it. How well people communicate and work with people is more important than being up on the latest programming language. is terrible. by integrating them into your work. at a minimum. How to Grow Professionally Assume responsibility in excess of your authority. A significant portion of everyone's energy should be devoted to recruitment. If you want to become a better programmer.5. I have several times heard interviewees say that the quality of the examination was one of their motivations for choosing a company. You should. as well as things as that help you personally. You can usually do this comfortably while working with a leader or a manager. This has the advantage that can uncover the interviewee that can present themselves well but can't really code---and there are many such people. who will know less but have a greater impact on your career. How to Evaluate Interviewees Evaluating potential employees is not given the energy it deserves. do it a little at a time. You may be able to recognize it by comparing notes after the interview. 3. and the human capacity for self-deception is astonishing. take responsibility for the schedule. You should also watch for the whiff of brimstone that is given off by difficult people. A bad hire. both the trivial technical kind. and the hard social kind. Play the role that you desire. like learning a new software system. designed to put the candidate under a great deal of stress. Interviewees will respect this. you will be able to quickly cover what they know and quickly retract from what they don't know to mark out the boundary. There are different interviewing styles. ask someone you admire how you can become like them.Intermediate consider the long term maintenance costs of both solutions. you should also evaluate their ability to learn. though this is rarely done. like a bad marriage. With practice. instigate the formation of consensus. give the candidate the equivalent of an oral examination on the technical skills for two hours. not where they worked last or what school they went to or some other inessential characteristic. you will have to do a thorough evaluation of the software before you buy it. In doing this. 3. which is far more important than what they know. You can also ask your boss. Plan ways to learn new skills.6. you will assume an unreasonable risk in buying it and you should decide against buying that particular product. I personally have not tried this technique. A reader has had good luck using a ‘take-home’ test for interviewees. but it sounds sensible. 30 .
you are talking to a programmer. It is possible that I have been extraordinarily lucky in the programmers that I have been able to work with. then finish strong with the good stuff. but not as grounded in creating technical things as we are. We make things. Here a programmer's experience and judgment can truly synergize with the fancy technology to make integration easy. There is no point in improving an algorithm when most of your time is spent making inefficient database calls. 3. and other gee-whiz stuff that most programmers know about but rarely use. However. interviewing is also a process of selling.Intermediate Finally. 31 . This implies that other people are different from us.7. An unfortunate amount of programming consists of getting systems to talk to each other and using very simple data structures to build a nice user interface. but they are not experts on making things. and that others will find the stereotype of the communication-challenged programmer the norm. so don't try to color the truth. I value non-engineers very highly. They are not as good at working together on teams as engineers are (there are no doubt exceptions. The three most important considerations for the potential computer science technique are: • • • Is it well encapsulated so that the risk to other systems is low and the overall increase in complexity and maintenance cost is low? Is the benefit startling (for example.8. data structures. this wonderful stuff is too complicated and generally unnecessary. but it would be dishonest for me to retract them until experience gives me a counter-example. In practice. for instance. They sell things and handle things and count things and manage things. This is worth bearing in mind when communicating with non-engineers. mathematics. I do apologize if anyone is offended by these beliefs. you should always understand the audience. the only issue is the risk of integration. then it would be criminal not to consider it. When is high technology the appropriate technology? When should you crack a book to get something other than a run-of-the-mill algorithm? It is sometimes useful to do this but it should be evaluated carefully. since the proposed technology has probably been well studied. You should be selling your company or project to the candidate. 3.)3 Their social skills are generally as good as or better than engineers in 3A number of readers have found this section arrogant and out of touch with their own experience. a factor of two in a mature system or a factor of ten in a new system)? Will you be able to test and evaluate it effectively? If a well-isolated algorithm that uses a slightly fancy algorithm can decrease hardware cost or increase performance by a factor of two across an entire system. How to Know When to Apply Fancy Computer Science There is a body of knowledge about algorithms. Start off with the bad stuff. One of the keys to arguing for such an approach is to show that the risk is really quite low. I am not being intentionally condescending. How to Talk to Non-Engineers Engineers and programmers in particular are generally recognized by popular culture as being different from other people. Non-engineers are smart.
With outsiders. Often non-engineers propose solutions that they think will make it easier on us out of kindness and a desire to do the right thing. we can often find a simple solution to a problem. use a sort of shorthand. Non-programmers can understand technical things but they do not have the thing that is so hard even for us---technical judgment. wastes their time. even worse. Because we have technical judgment and can usually understand business issues.) This represents a great opportunity to synergize with them. and. an abbreviated language that is effective because you will have much shared experience about technology in general and your product in particular. and how you can do it better. If you have the opportunity to meet with them often. the basic assumptions and goals do not need to be restated often. They do understand how technology works. Non-engineers may be too eager to please and they may be intimidated by you. You should assume that you will miscommunicate and watch carefully to find this miscommunication. This vocabulary create a wall between you and those that do not share it. If there is a problem in communication. and non-engineers like this about us. to bring clarity out of confusion. and then not stand behind their words. Try to get them to summarize or paraphrase what you are saying to make sure they understand. in terms of the clarity of your communication. You can often lead by example. They may not understand things you take for granted. (After all. they may say ‘yes’ without really meaning it to please you or because they are a little scared of you. 32 . especially when members of your own team are present. When talking to your team you will. and most conversation focuses on the details. I personally like Extreme Programming because it addresses this inefficiency. seek to alter your own practices before becoming frustrated with theirs. when in fact a much better overall solution exists which can only be seen by synergizing the outsiders view with your technical judgment. but they cannot understand why a certain approach would take three months and another one three days. you can leave a conversation with an outsider thinking that you understand each other when really there is a large misunderstanding. but their work does not always demand that they practice the kind of intimate. Just like us. programmers are anecdotally horrible at this kind of estimation as well.Intermediate non-team environments. precise communication and careful subdivisions of tasks that we do. it makes it easier to find the idea that is the best combination of cost and benefit. It takes some effort not to use this shorthand with those that don't have that shared experience. Since you take them for granted and don't repeat them. without thinking. it must be the other way around. spend a little bit of time asking if you you are communicating effectively. I love working with non-engineers. by marrying the estimation quickly to the idea. It provides great opportunities to learn and to teach. Engineers are trained to bring order out of chaos. With your team.
and have the customer satisfaction based on the original styling be so great that we make a lot of money. You should think twice before creating a new one. Unless you have some truly original needs and ideas. It is impossible to satisfy a vague requirement. something is impossible if either it cannot be grown from a simple system or it cannot be estimated. but not necessarily impossible. It makes the system tremendously powerful. It allows you to exercise your most creative and Promethean skills. But we should not let that blind us to the needs of the user. It is a joy to create a new language. How to Utilize Embedded Languages Embedding a programming language into a system has an almost erotic fascination to a programmer. Advanced 1. By this definition what is called research is impossible. How to Tell the Hard From the Impossible It is our job to do the hard and discern the impossible. such as ‘Build a system to compute an attractive hair style and color for a person. 1. but if it meshes with their culture they will not have to spend much time learning it. I and many other programmers have fallen into the trap of creating special purpose embedded languages. as it is in text editors. you will not succeed. would they prefer an applications programmers interface (API)? And what language will it be? Programmers don't want to learn a new language that is narrowly used. it will often become merely hard. A large volume of mere work is hard. so that initiates can use it and no one else has to. I fell into it twice. Technological Judgment 1. either from a scientific point of view or a software engineering point of view. This can be used to the extent that the intended audience can master the language. There already exist many languages designed specifically to be embedded languages. such as ‘Build a system that will compute the most attractive hair style and color for any person. why not use an existing language so that you can leverage the familiarity users already have with it? 33 . allow them to preview it and make changes.’ If there is not crisp definition of success. It then becomes your job to help the entrepreneur find a reasonable solution which is merely hard and gets most of what they wanted. use of the language can be made optional. The real question to ask oneself before embedding a language is: Does this work with or against the culture of my audience? If you intended audience is exclusively non-programmers. The best text editors in the world all have embedded languages.’ If the requirement can be made more crisp.2.Chapter 4. Of course. It is one of the most creative acts that can be performed. From the point of view of most working programmers.1. It makes the system into your friend. The distinction is not facetious because you may very well be asked to do what is practically impossible. how will it help? If your intended audience is exclusively programmers. A solution is merely hard when it can be confidently scheduled and the risks are understood.
Schedule pressure is the pressure to deliver something faster than it can be delivered and it is wasteful. Even if a core language is mandated and beyond your control. but psychology matters. and to some extent with a larger community. You can evolve to a new language/platform easily by rewriting each component individually. But of course. Perhaps because they project their own behavior onto us. things are never one-dimensional. this issue is dictated by pointy-haired bosses who are making a political decision. Some of these effects may only be psychological. precludes choice on the part of the individual. Programming languages should really be called notations in that learning one is not at all as difficult as learning a natural language. If a language is to be embedded (and you should always consider it!) the choice of language will depend a lot on the culture of the users. but the effect is very small. However. The people who task programmers do not fully appreciate what a strong work ethic we have and how much fun it is to be a programmer.1. they believe that asking for it sooner will make us work harder to get it there sooner. 34 .Advanced 1. Its possible that some of the modules are actually up-to-date. and in so doing make work more interesting. it is often the case that tools and other programs can and should be written in a different language. that the less accepted tool is best. unhealthy. and is healthy up to a point. Compromising Wisely 2. One tends to think of a large system that has components in three or four languages as a messy hodgepodge. In the end the costs of language tyranny outweigh any advantage that it provides. To beginners and to some outsiders ``learning a new language'' seems a daunting task. It is good because it reflects a financial reality. Generally. 2. How to Fight Schedule Pressure Time-to-market pressure is the pressure to deliver a good product quickly. I personally believe this the most wasteful and erroneous common practice you are likely to encounter. Most working programmers have very little control of the language they will use. rather than a technological decision. Schedule pressure exists for several reasons. and lack the courage to promote an unconventional tool even when they know. This is probably actually true. but after you have three under your belt it's really just a question of becoming familiar with the available libraries. but I argue that such a system is in many cases stronger than a one-language system in several ways: • • • There is necessarily loose coupling between the components that are written in different notations (though maybe not clean interfaces). No doubt they are serving what they perceive to be the best interests of the project or company. One should take advantage of this to serve your company or project by using the best language for the job. Often managers are driven by the need to be able to hire programmers with experience in a given language. often with firsthand knowledge. and all too common. Choosing Languages The solitary programmer that loves her work (a hacker) can choose the best language for the task. In other cases the very real benefit of unity among the team. and must be respected for that.3.
Additionally. The user has their own job. Not being able to see it. Your confidence in your own ideas about this should vary. The more time you spend with users the better you will be able to understand what will really be successful. The key to fighting schedule pressure is simply to turn it into time-to-market pressure. but they may not have the vision to do this. You should try to test your ideas against them as much as you can. working harmoniously together give the best chance of formulating the correct vision. Producing an honest. Because the user is not as intimately involved in the creation of your product as you are. If you intend to be a consultant. understandable estimate of all the labor involved is the best way to do this. Programmers cannot be hoodwinked. The user can't have a vision that represents the complete body of your product users. a programmer should never say ‘no’. and not be able to create it themselves. they behave a little differently: • • • The user generally makes short pronouncements. detailed. You should eat and drink with them if you can. I suggest you choose 35 . Setting an unrealistic schedule will also be painfully obvious to everyone. I hope that every reader will be lucky enough to use it. In a sense. Programmers' hard work will be visible. they will mainly think of small improvements in your product. they have no visibility into what it really takes to produce software. Programmers are trained to design and create. You must guard against both arrogance and false modesty in terms of knowing what the customer really wants. How to Understand the User It is your duty to understand the user. but rather to say ‘What will you give up to get that thing you want?’ The effect of producing clear estimates will be to increase the respect for programmers. Extreme Programming amplifies this and builds a process around it. Market researchers are trained to figure out what people want. The key insight that the estimate must make plain is that labor is an almost incompressible fluid. It is disrespectful and demoralizing to ask them to do something unrealistic. not big improvements. This is how other professionals behave.Advanced and the damage is very great. and most of all. Guy Kawasaki[Rules] has emphasized the importance of watching what your users do in addition to listening to them. 2. the only thing they can do is see time-to-market pressure and fuss at programmers about it. It has the added advantage of allowing good management decisions to be made about possible functionality tradeoffs. not what they say they want. You can't pack more into a span of time anymore than you can pack more water into a container over and above that container's volume. These two kinds of people. better to propose it to them and get them to agree that your proposal is what they really want before you begin. I believe contractors and consultants often have tremendous problems getting their clients to clarify in their own minds what they really want. It is your duty to give them what they really want. It is however. The way to do this to give visibility into the relationship between the available labor and the product. and to help your boss understand the user.2. or two modes of thought in the same person.
Serving Your Team 3. with one caveat: you cannot be appreciated without visibility into your work. If there is never any failure. but often times your perception of what you need to do will differ considerably from your boss's. 3. you should be as gentle as you can with them while not treating them as though they had succeeded. do to happenstance or personal habits. 2. but I think people are generally moderately fair. To get a pay raise. Working from home a lot or being geographically separated from your team and boss makes this especially difficult. Try to get each team member to buy in and be well motivated. You must try to get them well-motivated 36 . Stretch them not by giving them more work. This sounds trite. If there are not occasional failures. or every other project. Sometimes. Ask them explicitly what you need to do to get promoted.3. but you should know what everybody desires. and try to do it. but not overburden them. You should allow people (including yourself) to fail occasionally and should plan for some failure in your schedule. Also this will pin your boss down in some ways. but by giving them a new skill or better yet a new role to play on the team. When someone fails. try to stretch them in both a way that they suggest and a way that you think will be good for them. makes me stronger. act out that role first. You should usually talk to them about the way they are being stretched. How to Develop Talent Nietschze exaggerated when he said[Stronger]: What does not destroy me. To get promoted to a title. we can't all be in the top 10%! However. find out what is expected of that title and do that. You should challenge your team. negotiate armed with information. someone will not be noticed much. they will be well motivated. I have seem some people who were seriously unappreciated. One cannot expect everyone's evaluation to perfectly match reality at all times. You can't give up on someone who is intentionally not carrying their share of the load because of low morale or dissatisfaction and just let them be slack.Advanced your clients based on their clear-headedness as well as their pocketbooks. You may have to leave them dissatisfied. talk to your boss about it. How to Get a Promotion To be promoted to a role. there can be no sense of adventure. If they buy in to it. Your greatest responsibility is to your team.1. you are not taking enough risks. You should know each of them well. On each project. Most programmers probably have an exaggerated sense of their relative abilities in some ways--after all. If you feel like you are past due for a promotion. Ask each of them explicitly what they need to be well-motivated if they are not.
If you are very strong. your tribe would lose some important benefits.2. However. Leadership and communication skills are more important than technical skills. simply disagree openly and accept the consensus. This is perfectly natural and nobody will be bothered by it as long as everyone works hard. simply assert that you have to make a decision and that is what your decision is. but carefully delineated. If you did this you would in fact make more progress in the short term. Praise should be public and criticism private. This occasionally means allowing your teammates to be wrong. How to Get the Most From Your Teammates To get the most from your teammates. If there is an individual who will not consent after you have discussed the issues from all appropriate sides. even if you believe with great confidence it is the wrong thing to do. You cannot allow someone who is intentionally working below their level to remain on the team. if it does not harm the project too much. This creates a strange situation. corny stuff like logoized clothing and parties are good. As a leader. or like you're being forced into it. but not as good as personal respect. You can often give the stronger team members challenging. Make it clear to the strong members of your team that you think they are strong by saying so in public. It will often be true that you could move faster if your weak programmers would just get out of the way. you can't ask for more than you give yourself in this respect. simply state that you disagree but think the consensus of the team is more important. It is an odd fact that is not reflected in salaries that a good programmer is more productive than 10 bad programmers. How to Choose What to Work On You balance your personal needs against the needs of the team in choosing what aspect of a project to work on. Team spirit is created when people make sacrifices for the team and think in terms of the good of the team before their own personal good. If there 37 . develop a good team spirit and try to keep every individual both personally challenged and personally engaged. When this happens. Don't sound hurt. and the ability to recover from the loss of the strong members. Don't insist that they go through with their initial plan if they do backtrack. fire them. The strong must be gentle in this regard and consider the issue from all angles. since it is not fair to the team. keep this up. 3. To develop team spirit. don't agree. One of the keys to team leadership is to facilitate consensus so that everyone has buy in. but try to find a way to stretch yourself not by taking on more work but by exercising a new skill. As long as you have the patience.Advanced and productive. based on consensus. If everyone respects everyone else. You should do what you are best at. When your patience is exhausted. 3.3. That is. namely the training of the weaker members. The strong members of the team will naturally have more difficult tasks than the weak members of the team. and do it as early as possible in the project to decrease risk. the spreading of tribal knowledge. tasks. This will often cause them to backtrack. take on the hardest or riskiest task. nobody will want to let anybody down. you must let some of your team do things their own way.
5. At a minimum assign a team of two to do the work and promote healthy teamwork to get the task done. what they think would create team spirit and make for an effective team. Especially praise those who disagree with you when they are praiseworthy. A prototype is always powerful but in software it is far superior to any written description. specialization is a very useful productivity tool when not overused. These tasks can really hurt the morale of those that have to do them. Ask your team. switch as quickly as you can and recognize the persons who were right. This is impossible since the productivity of individuals varies so widely. However. This is part of the reason that a high-performing team should not be broken up. with one exception: sometimes growth or the correction of a fault can't be praised without drawing embarrassing attention to the original fault. Wherever possible. but under no circumstances allow them to do it alone. Try to find some way to get the computer to do the task for you or to help your teammates do this. Individually recruit key supporters for your project. create and communicate a vision that demonstrates real value to the organization as a whole. Praise in public and criticize in private. Just as a composer considers the timbre of the instrument that will play a part or the coach of an athletic team considers the strengths of each player. don't tell. Who has particular knowledge about a component also constantly changes and can have an order of magnitude effect on performance. If possible. How to Gather Support for a Project To gather support for a project. This should be done early. construct a prototype or a mockup to demonstrate your ideas. so that growth should be praised in private. 3. both as a group and individually. Attempt to let others share in your vision creation. 3. show. Working for a week on a program to do a task that will take a week to do by hand has the great advantage of being more educational and sometimes more repeatable. If all else fails.6. Praise frequently rather than lavishly.Advanced is a way to judge if your decision will be wrong or if it is later shown to be wrong. apologize to those who have to do the boring task. 38 .4. 3. There is a certain danger in this given that people will become bored as they build upon their strengths and never improve their weaknesses or learn new skills. The best technique for dealing with this is to invoke or promote Larry Wall's programmer's virtue of Laziness. How to Divide Problems It's fun to take a software project and divide it up into tasks that will be performed by individuals. Sometimes managers like to think that an estimate can be made without consideration of the individuals that will perform the work. This gives them a reason to support you and gives you the benefit of their ideas. How to Handle Boring Tasks Sometimes it is not possible to avoid boring tasks that are critical to the success of the company or the project. the experienced team leader will not usually be able to separate the division of the project into tasks from the team members to which they will be assigned.
and so on. but merely an unfinished bridge. there is never a baby bridge. it dies and its body feeds other organisms. We have the luxury of treating software like that. How to Communicate Well To communicate well. It is a skill unto itself. you have to recognize how hard it is. This approach is captured in spiral development. you must visualize the end result and communicate it in a way that the engineering team can get excited about. But you must also communicate to them a path that goes from wherever they are now to where they want to be with no large leaps. Pruning off weak limbs is healthful. They speak poorly and write poorly. A reader. One of the programmer's jobs is to balance the immediate payoff against future payoff by wisely choosing a growth path expressed in milestones. The embryo grows. Rob Hafernik. Nothing is as bad as the special horror of getting down to the end and discovering that the whole thing just isn't going to work (look at the recent debacle of the Voter News System). Eventually it bears fruit. They do not work hard at understanding you. it cannot be dead at one point and resurrected later. The advanced programmer has the triple responsibility of growing software. They are often overworked 39 . In the ultra-competitive environment of business. it is best if the milestones can be released and make money as early as possible. It's not just systems. A bridge is not like that. data models. The tree must stay alive the whole time. because it allows us to make useful progress before we have a perfect mental image. It becomes larger. Bridges are a lot simpler than software. To which one can only reply Fiat lux! 3. it can only be evolved from a simple system to a complex system in a series of intentional steps. It is your job to take a germ of an idea and build a path that takes it as smoothly as possible into a useful artifact. It's absolutely vital as you work on a large system to have measurable progress toward intermediate goals. complex system can be implemented from scratch. To do this. teams. It is good to think of software as growing. Later.Advanced 3. but algorithms.8. It looks more like the adult and has more of the uses. even if they are far away from a well-designed endpoint. sent in this comment on this section that I can do no better than to quote in full: I think you under-emphasize the importance here. The programmer must design a finished system that can be delivered and used. Milestones that are never too far apart are used to mark progress along the path. It is made harder by the fact that the persons with whom you have to communicate are flawed.7. How to Grow a System The seed of a tree contains the idea of the adult but does not fully realize the form and potency of the adult. I would even go further and state it as a law of nature: no large. and persons. You must design a growth path that ends in the finished system. We can get feedback from users and use that to correct the growth. But the advanced programmer must do more. user interfaces.
I think a lot of times we are not patient enough in this kind of difficult communication. There is nothing worse than postponing action when a milestone slips. even if the only action is to inform everyone. How to Tell People Things They Don't Want to Hear You will often have to tell people things that will make them uncomfortable. you are telling them as early as possible so they will be well-informed. The second best way is to appeal to them for help with the problem. The programmer brings order out of chaos. I haven't mastered communicating well.’ The conscientious programmer hates to say this. You will want your team's input on both where you stand and what can be done about it. at least in spirit. somewhat focused on their own work rather than the larger issues you may be addressing. One of the advantages of taking classes and practicing writing. and worse. In doing this. This can be done in a strawman or white-paper format or just verbally. ‘The schedule will have to slip. and the team will have to face the consequences with you. It also means a story of religious significance that explains the universe and mankind's relationship to it. if not physically. Programmers are the part of the bridge between ideas and reality that rests on reality. Entrepreneurs who are not programmers need programmers to provide leadership in some ways. If you have invested energy in there preparation. because she has a unique power and therefore a unique responsibility.9. The advanced programmer must be prepared to accept this. For that reason. and. The programmer is a social animal whose survival depends on communication with her team. How to Deal with Managerial Myths The word myth sometimes means fiction. Managers tend to forget what they learned as programmers and believe in certain myths. and listening is that if you become good at it you can more readily see where problems lie and how to correct them. If there is a danger that you won't be believed. public speaking. It would be as rude and unsuccessful to try to convince them these myths are false as to try to disillusion a devoutly religious person of their beliefs. 3. Even if nothing can be done about the problem. One interesting way to do this is to initiate a proposal of some kind outside the team. The best way to tell someone about a problem is to offer a solution at the same time. I try to speak verbally. you should recognize these beliefs as myths: 40 . The advanced programmer is a social animal whose satisfaction depends on communication with people outside her team. Remember that you are doing this for a reason. hand people a white-paper (on real paper.10. it is better to do it as a team. It also exposes you to criticism. but must say it as early as possible. 3. You should not be disheartened if your ideas are not immediately accepted. but what I'm currently trying is what I think of as a fourpronged approach: After I have my ideas in order and am fully prepared. One of the most unpleasant and common things you will have to say is. as well as electronically) show them a demo.Advanced or bored. you should gather some support for your assertion. But it has a deeper connotation. no one will think poorly of you for it. This leadership has the tremendous advantage of setting the terms of the debate. at a minimum. rejection and neglect. and then patiently repeat this process.
(The cost of communication with the new persons is almost always more taxing than helpful. If you are really sure that it is stupid. tell your people to do the same thing and tell them to ignore what anybody else tells them. 41 .) Resources can be added to a late project to speed it. but don't feel bad if you have no success and don't damage your reputation by confronting these myths belligerently. When you have it you should ignore the chaos completely and carry on as if nothing is happening. ipos. not good.) Programmers' productivity can be measured in terms of some simple metric.11. (They want it. (Programmers vary by an order of magnitude. such as layoffs. and impede if they are bad. If you are not a programmer. (It is not even theoretically possible. Organizational chaos is a great opportunity for programmers to exercise their magic power. This course of action is the best for you personally. 3. These are unsettling to everyone. but perhaps a little less unsettling to the programmer whose personal self-esteem is founded in capacity rather than in position. More commonly. Engineers have the power to create and sustain. If you are a leader. like lines of code. and so on. get fired. Each of these myths reinforces the manager's idea that they have some actual control over what is going on. lines of code are bad.) Programmers can be equated.) It is possible to estimate software development reliably. just as engineers typically cannot sell a product or manage a business effectively. in a typical software company. buyouts. but they don't want you to spend any time on it. The truth is that managers facilitate if they are good. new hirings. it is best to smile and nod until they go away and then carry on doing what you know is best for the company. How to Deal with Temporary Organizational Chaos There are often brief periods of great organizational chaos. and is the best for your company or project. Non-engineers can order people around but. can create and sustain nothing without engineers. you can try to explain these things. (If succinctness is power. You may. firings. I've saved this for last because it is a deep tribal secret. some stressed-out person who does not have the magic power will come into your cube and tell you to do something stupid. but if that happens you can probably get a new job because of the magic power.Advanced • • • • • More documentation is always better. This power is proof against almost all of the problems associated with temporary organizational mayhem. of course.) If you have an opportunity. please stop reading now.
build An adjective describing a choice between spending money for software or writing it your self. This contrast with formal diagram methodologies. divide and conquer A technique of top-down design and. boss The person or entity that gives you tasks. These do not necessarily have a standardized meaning to other people. of debugging that is the subdivision of a problem or a mystery into progressively smaller problems or mysteries. garbage Allocated memory that no longer has any useful meaning. communication languages A language designed primarily for standardization rather than execution.Glossary This is a glossary of terms as used in this essay. importantly. bottleneck The most important limitation in the performance of a system. Raymond has compiled a massive and informative glossary[HackerDict] that rather surprisingly can pleasurably be read cover-to-cover once you can appreciate a fraction of it. informal style of making diagrams consiting of boxes and arrows drawn between those boxes to show the relationships. busines A group of people organized for making money. buy vs. entrepreneur The initiator of projects. such as UML. In some cases this may be the public at large. garbage collector A system for recycling garbage. company A group of people organized reflecting the modern workspace. Eric S. Extreme Programming A style of programming emphasizing communication with the 42 . boxes and arrows A loose. A constriction that limits performance.
Mere work can be estimated easily. information hiding A design principle that seeks to keep things independent and decoupled by using interfaces that expose as little information as possible. printlining The insertion of statements into a program on a strictly temporary basis that output information about the execution of the program for the purpose of debugging. object-oriented programming An programming style emphasizing the the management of state inside objects. heap allocated Memory can be said to be heap allocated whenever the mechanism for freeing it is complicated. logging The practice of writing a program so that it can produce a configurable output log describing its execution. memory leak The unwanted collection of references to objects that prevents garbage collection (or a bug in the garbage collector or memory management system!) that causes the program to gradually increase its memory demands over time. mere work Work that requires little creativity and entails little risk. lingua franca A language so popular as to be a de facto standard for its field.Glossary customer and automated testing. hitting the wall To run out of a specific resource causing performance to degrade sharply rather than gradually. programming notation A synonym for programming language that emphasizes the mathematical nature of programming language and their rela- 43 . as French was for international diplomacy at one time. low-hanging fruit Big improvements that cost little. master A unique piece of information from which all cached copies are derived that serves as the official definition of that data.
as often as not. strawman A document meant to be the starting point of a technical discussion. unk-unk Slang for unknown-unknown. will never materialize into anything solid. woodman. A strawman may lead to a stickman. 44 . tribe A group of people you share cultural affinity and loyalty with.Glossary tive simplicity compared to natural languages. vapor Illusory and often deceptive promises of software that is not yet for sale and. scroll blindness The effect of being unable to find information you need because it is buried in too much other. such as one on a wall. Problems that cannot presently even be conceptualized that will steal time away from the project and wreck the schedule. wall-clock Actual time as measured by a physical clock. speculative programming Producing a feature before it is really known if that feature will be useful. tinman. ironman. etc. white-paper An informative document that is often meant to explain or sell a product or idea to an audience different than the programmers of that product or idea. less interesting information. as opposed to CPU time.
Microsoft Press. Addison-Wesley.com/articles.html. 0201616416. http://catb.utexas. Redmond.Appendix A. Extreme Programming Explained: Embrace Change. Michelle Moreno. 2003. 2000. 1984.PDF . 0201710919.html. The to New Become Hacker a Hacker. 1999. 2000.edu/users/EWD/ewd09xx/EWD913. but especially "Beating the Averages". HarperBusiness. AddisonWesley. Microsoft Press.utexas. Wash. Addison-Wesley. Dijkstra. 2003. David Thomas. Edsger W. 1993. [XP99] Kent Beck.catb. Wash. Code Complete. Dictionary. 1999. Gap. [HackDict] Eric S. 1986. 1889. Articles on his website: http://www. [CodeC93] Steve McConnell. 020161622X. and Ward Cunningham. Twilight of the Idols. Raymond. Bibliography Bibliography Books [Rules00] Guy Kawasaki. [Prag99] Andrew Hunt. Web Sites [PGSite] Paul Graham. Rules for Revolutionaries: The Capitalist Manifesto for Creating and Marketing New Products and Services. The Pragmatic Programmer: From Journeyman to Master. [ExpCS] [Knife] Edsger W. [RDev96] Steve McConnell. 2002.cs. Raymond. All of them.. and Gary Kawasaki. section 8. Planning Extreme Programming. [PlanXP00] Kent Beck and Martin Fowler. [Stronger] Friedrich Nietzsche. 1996.org/esr/jargon/jargon. [Hacker] Eric S. Dijkstra. Science?. How http://www. Rapid Development: Taming Wild Software Schedules. On a http://www.edu/users/EWD/ewd09xx/EWD988a.org/~esr/faqs/hacker-howto. Cultural 45 .PDF.html.cs.paulgraham. "Maxims and Arrows". How Experimental is Computing http://www. Redmond.
I will consider it and may include it in subsequent versions that I produce. and/or Criticism or commentary in-lined into the text. The ability to build into different formats. Although not a textbook. Request for Feedback or Extension Please send me any comments you may have on this essay. I have placed this essay under the GNU Free Documentation License. 2003) B.2. This license is not specifically designed for essays. After this article was mentioned on Slashdot in 2003.1. Read in the year 2000 and first published electronically at Samizdat Press(http://Samizdat. Read B. I appreciate them all. even if only on a subsection-by-subsection basis. I hope this essay is a short and pleasant read. Essays are usually intended to be coherent and convincing arguments that are writtien from a single point of view in a single voice.Appendix B. If you inform me of your work. subject to the provisions of the License.edu) in 2002. History (As Of May. I consider all suggestions. you are encouraged to expand upon this essay as you see fit. Thank you. I would be joyous if it were extended in the following ways: • • • • • The addition of a comprehensive reading list to each section. but hope springs eternal. but the following people ei- 46 . I also hope that it is instructive.mines. If so inclined. such as palm formats and better HTML. It may be arrogance to imagine that this document is worthy of extension. You may of course produce your own versions of this document without my knowledge.com. as explained in the License. many of which have already improved this essay. There was a lot of duplication. The addition of more and better sections. It is dedicated to the programmers of Hire. Original Version The original version of this document was begun by Robert L. subject to the provisions of the License. it is broken into many small sections to which new sections can be freely added. Robert L. Translation into other languages. about 75 people sent me email with suggestions and errata.
In 1987 he received a BA in Computer Science from Rice University. and Wayne Allen.com. B. 2003) ther made major suggestions or were the first to find a mistake that I fixed: Morgan McGuire. Prior to that he founded 4R Technology. He has been a paid programmer since the age of 16. Mark Howe. where he has worked for four years. Ninja Programmer (145252) at Slashdot. Maksim Ioffe. Rob received a PhD from the University of Texas at Austin in 1995 in Computer Science related to database theory. Andrew Wu. who encouraged me to initiate this. Finally I would like to thank Christina Vallery. Read lives in Austin. Zed A. with his wife and two children. Original Author's Bio Robert L. Shaw.3. Rob Hafernik.History (As Of May. He is currently a Principal Engineer at Hire. which made a scanner-based image analysis quality control tool for the paper industry. Tom Moertel. Ben Vierck. David Mason. Steve Benz. David Jeschke. 47 . whose editing and proofreading greatly improved the second draft. Brian Grayson. Texas. Pieter Pareit. and Tom Corcoran.
We recommend this License principally for works whose purpose is instruction or reference. PREAMBLE The purpose of this License is to make a manual.2002 Free Software Foundation. with or without modifying it. and is addressed as "you". while not being considered responsible for modifications made by others. MA 02111-1307 USA Everyone is permitted to copy and distribute verbatim copies of this license document. GNU Documentation License Version 1. because free software needs free documentation: a free program should come with manuals providing the same freedoms that the software does. A "Secondary Section" is a named appendix or a front-matter section of the Document that deals exclusively with the relationship of the publishers or authors of the Document to the Document's overall subject (or to related matters) and contains nothing that could fall directly within that overall subject. Inc. (Thus. royalty-free license. which means that derivative works of the document must themselves be free in the same sense. Suite 330. This License is a kind of "copyleft". Such a notice grants a world-wide. if the Document is in part a textbook of mathematics. APPLICABILITY AND DEFINITIONS This License applies to any manual or other work. that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. a Secondary Section may not explain any mathematics.2001. A "Modified Version" of the Document means any work containing the Document or a portion of it. textbook. or with modifications and/or translated into another language. either commercially or noncommercially. Secondarily. 2. The "Document". in any medium. Any member of the public is a licensee. which is a copyleft license designed for free software. or other functional and useful document "free" in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it. or of legal.) The relationship could be a matter of historical connection with the subject or with related matters. regardless of subject matter or whether it is published as a printed book. this License preserves for the author and publisher a way to get credit for their work. 48 . But this License is not limited to software manuals. It complements the GNU General Public License. but changing it is not allowed. philosophical. You accept the license if you copy. 59 Temple Place. November 2002 Free Copyright (C) 2000. modify or distribute the work in a way requiring permission under copyright law. unlimited in duration. below. Boston. either copied verbatim. refers to any such manual or work.2. it can be used for any textual work. to use that work under the conditions stated herein. We have designed this License in order to use it for manuals for free software.Appendix C. 1. commercial. ethical or political position regarding them.
Texinfo input format. plus such following pages as are needed to hold. and that you add no other conditions whatsoever to those of this License. You may not use technical measures to obstruct or control the 49 . (Here XYZ stands for a specific section name mentioned below. A "Transparent" copy of the Document means a machine-readable copy. PostScript or PDF produced by some word processors for output purposes only. The Document may contain zero Invariant Sections. and the machine-generated HTML. and the license notice saying this License applies to the Document are reproduced in all copies. SGML or XML for which the DTD and/or processing tools are not generally available. and a Back-Cover Text may be at most 25 words. VERBATIM COPYING You may copy and distribute the Document in any medium. LaTeX input format. has been arranged to thwart or discourage subsequent modification by readers is not Transparent. the title page itself. If the Document does not identify any Invariant Sections then there are none. the copyright notices. A section "Entitled XYZ" means a named subunit of the Document whose title either is precisely XYZ or contains XYZ in parentheses following text that translates XYZ in another language. A FrontCover Text may be at most 5 words. and that is suitable for input to text formatters or for automatic translation to a variety of formats suitable for input to text formatters. Examples of suitable formats for Transparent copies include plain ASCII without markup. The "Cover Texts" are certain short passages of text that are listed. provided that this License. SGML or XML using a publicly available DTD. The Document may include Warranty Disclaimers next to the notice which states that this License applies to the Document. "Endorsements". If a section does not fit the above definition of Secondary then it is not allowed to be designated as Invariant. A copy that is not "Transparent" is called "Opaque". in the notice that says that the Document is released under this License. The "Title Page" means. "Title Page" means the text near the most prominent appearance of the work's title. Opaque formats include proprietary formats that can be read and edited only by proprietary word processors. PostScript or PDF designed for human modification. XCF and JPG. or "History". but only as regards disclaiming warranties: any other implication that these Warranty Disclaimers may have is void and has no effect on the meaning of this License. the material this License requires to appear in the title page. such as "Acknowledgements". An image format is not Transparent if used for any substantial amount of text. as being those of Invariant Sections. either commercially or ncommernocially. or absence of markup. A copy made in an otherwise Transparent file format whose markup. "Dedications".GNU Free Documentation License The "Invariant Sections" are certain Secondary Sections whose titles are designated. that is suitable for revising the document straightforwardly with generic text editors or (for images composed of pixels) generic paint programs or (for drawings) some widely available drawing editor.) To "Preserve the Title" of such a section when you modify the Document means that it remains a section "Entitled XYZ" according to this definition. represented in a format whose specification is available to the general public. For works in formats which do not have any title page as such. as Front-Cover Texts or BackCover Texts. in the notice that says that the Document is released under this License. 3. for a printed book. and standardconforming simple HTML. These Warranty Disclaimers are considered to be included by reference in this License. legibly. Examples of transparent image formats include PNG. preceding the beginning of the body of the text.
Use in the Title Page (and on the covers. you may accept compensation in exchange for copies. to ensure that this Transparent copy will remain thus accessible at the stated location until at least one year after the last time you distribute an Opaque copy (directly or through your agents or retailers) of that edition to the public. 5. as long as they preserve the title of the Document and satisfy these conditions. You may also lend copies. be listed in the History section of the Document). as authors. you must do these things in the Modified Version: A. if there were any. List on the Title Page. 50 . under the same conditions stated above. and Back-Cover Texts on the back cover. and the Document's license notice requires Cover Texts. Both covers must also clearly and legibly identify you as the publisher of these copies. However. Copying with changes limited to the covers. COPYING IN QUANTITY If you publish printed copies (or copies in media that commonly have printed covers) of the Document. provided that you release the Modified Version under precisely this License. you should put the first ones listed (as many as fit reasonably) on the actual cover. you must enclose the copies in covers that carry. You may use the same title as a previous version if the original publisher of that version gives permission. you must either include a machine-readable Transparent copy along with each Opaque copy. If the required texts for either cover are too voluminous to fit legibly. or state in or with each Opaque copy a computer-network location from which the general network-using public has access to download using public-standard network protocols a complete Transparent copy of the Document. and continue the rest onto adjacent pages. You may add other material on the covers in addition. MODIFICATIONS You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and 3 above. numbering more than 100. 4. to give them a chance to provide you with an updated version of the Document. The front cover must present the full title with all words of the title equally prominent and visible. if any) a title distinct from that of the Document. free of added material. If you publish or distribute Opaque copies of the Document numbering more than 100. It is requested. but not required. all these Cover Texts: FrontCover Texts on the front cover. and from those of previous versions (which should. when you begin distribution of Opaque copies in quantity. clearly and legibly. If you use the latter option. with the Modified Version filling the role of the Document. one or more persons or entities responsible for authorship B. If you distribute a large enough number of copies you must also follow the conditions in section 3. that you contact the authors of the Document well before redistributing any large number of copies. can be treated as verbatim copying in other respects. and you may publicly display copies. In addition.GNU Free Documentation License reading or further copying of the copies you make or distribute. you must take reasonably prudent steps. thus licensing distribution and modification of the Modified Version to whoever possesses a copy of it.
unless they release you from this requirement. These may be placed in the "History" section. Preserve the Title of the section. L. Preserve the section Entitled "History". For any section Entitled "Acknowledgements" or "Dedications". if it has fewer than five).GNU Free Documentation License of the modifications in the Modified Version. year. in the form shown in the Addendum below. F. You may omit a network location for a work that was published at least four years before the Document itself. D. Include an unaltered copy of this License. Preserve in that license notice the full lists of Invariant Sections and required Cover Texts given in the Document's license notice. N. Such a section may not be included in the Modified Version. you may at your option designate some or all of these sections as invariant. To do this. Preserve its Title. H. I. and preserve in the section all the substance and tone of each of the contributor acknowledgements and/or dedications given therein. Preserve the network location. together with at least five of the principal authors of the Document (all of its principal authors. a license notice giving the public permission to use the Modified Version under the terms of this License. E. Do not retitle any existing section to be Entitled "Endorsements" or to conflict in title with any Invariant Section. C. Add an appropriate copyright notice for your modifications adjacent to the other copyright notices. and add to it an item stating at least the title. given in the Document for public access to a Transparent copy of the Document. O. add their titles to the list of Invariant Sections in the Modified Version's license notice. authors. These titles must be distinct from any other sec- 51 . immediately after the copyright notices. G. year. if any. or if the original publisher of the version it refers to gives permission. M. Preserve any Warranty Disclaimers. as the publisher. K. new authors. If there is no section Entitled "History" in the Document. Include. and publisher of the Modified Version as given on the Title Page. and likewise the network locations given in the Document for previous versions it was based on. unaltered in their text and in their titles. If the Modified Version includes new front-matter sections or appendices that qualify as Secondary Sections and contain no material copied from the Document. Delete any section Entitled "Endorsements". and publisher of the Document as given on its Title Page. Preserve all the Invariant Sections of the Document. Preserve all the copyright notices of the Document. create one stating the title. then add an item describing the Modified Version as stated in the previous sentence. Section numbers or the equivalent are not considered part of the section titles. J. State on the Title page the name of the publisher of the Modified Version.
provided that you follow the rules of this License for verbatim copying of each of the documents in all other respects. The author(s) and publisher(s) of the Document do not by this License give permission to use their names for publicity for or to assert or imply endorsement of any Modified Version. COMBINING DOCUMENTS You may combine the Document with other documents released under this License. and any sections Entitled "Dedications". and distribute it individually under this License. you may not add another. forming one section Entitled "History". or else a unique number. under the terms defined in section 4 above for modified versions. and a passage of up to 25 words as a Back-Cover Text. 8. and follow this License in all other respects regarding verbatim copying of that document. Make the same adjustment to the section titles in the list of Invariant Sections in the license notice of the combined work. the name of the original author or publisher of that section if known. and replace the individual copies of this License in the various documents with a single copy that is included in the collection. on explicit permission from the previous publisher that added the old one. and that you preserve all their Warranty Disclaimers. If there are multiple Invariant Sections with the same name but different contents. You may add a section Entitled "Endorsements". You may extract a single document from such a collection. provided that you include in the combination all of the Invariant Sections of all of the original documents. and multiple identical Invariant Sections may be replaced with a single copy. in parentheses. make the title of each such section unique by adding at the end of it. You may add a passage of up to five words as a Front-Cover Text. likewise combine any sections Entitled Ac"knowledgements". previously added by you or by arrangement made by the same entity you are acting on behalf of. but you may replace the old one. to the end of the list of Cover Texts in the Modified Version. Only one passage of Front-Cover Text and one of Back-Cover Text may be added by (or through arrangements made by) any one entity. The combined work need only contain one copy of this License. You must delete all sections Entitled "Endorsements".GNU Free Documentation License tion titles. AGGREGATION WITH INDEPENDENT 52 . 7. statements of peer review or that the text has been approved by an organization as the authoritative definition of a standard. COLLECTIONS OF DOCUMENTS You may make a collection consisting of the Document and other documents released under this License. 6. provided you insert a copy of this License into the extracted document. unmodified. provided it contains nothing but endorsements of your Modified Version by various parties--for example. In the combination. and list them all as Invariant Sections of your combined work in its license notice. you must combine any sections Entitled "History" in the various original documents. If the Document already includes a cover text for the same cover.
You may include a translation of this License. the original version will prevail. If the Document specifies that a particular numbered version of this License "or any later version" applies to it. sublicense. in or on a volume of a storage or distribution medium.gnu. is called an "aggregate" if the copyright resulting from the compilation is not used to limit the legal rights of the compilation's users beyond what the individual works permit. When the Document is included in an aggregate. or rights. TERMINATION You may not copy. See http://www. 10. Otherwise they must appear on printed covers that bracket the whole aggregate.org/copyleft/. TRANSLATION Translation is considered a kind of modification. this License does not apply to the other works in the aggregate which are not themselves derivative works of the Document. Any other attempt to copy. from you under this License will not have their licenses terminated so long as such parties remain in full compliance. Such new versions will be similar in spirit to the present version. or the electronic equivalent of covers if the Document is in electronic form. and any Warranty Disclaimers. but you may include translations of some or all Invariant Sections in addition to the original versions of these Invariant Sections. However. Each version of the License is given a distinguishing version number.GNU Free Documentation License WORKS A compilation of the Document or its derivatives with other separate and independent documents or works. so you may distribute translations of the Document under the terms of section 4. revised versions of the GNU Free Documentation License from time to time. or "History". If the Cover Text requirement of section 3 is applicable to these copies of the Document. FUTURE REVISIONS OF THIS LICENSE The Free Software Foundation may publish new. If a section in the Document is Entitled "Acknowledgements". the Document's Cover Texts may be placed on covers that bracket the Document within the aggregate. 9. In case of a disagreement between the translation and the original version of this License or a notice or disclaimer. provided that you also include the original English version of this License and the original versions of those notices and disclaimers. Replacing Invariant Sections with translations requires special permission from their copyright holders. but may differ in detail to address new problems or concerns. or distribute the Document except as expressly provided for under this License. "Dedications". and will automatically terminate your rights under this License. modify. then if the Document is less than one half of the entire aggregate. parties who have received copies. you have the 53 . 11. sublicense or distribute the Document is void. and all the license notices in the Document. the requirement (section 4) to Preserve its Title (section 1) will typically require changing the actual title. modify.
replace the "with. with the Front-Cover Texts being LIST. If your document contains nontrivial examples of program code.Texts. you may choose any version ever published (not as a draft) by the Free Software Foundation. such as the GNU General Public License. or some other combination of the three. If the Document does not specify a version number of this License. merge those two alternatives to suit the situation. Front-Cover Texts and Back-Cover Texts. If you have Invariant Sections. A copy of the license is included in the section entitled "GNU Free Documentation License". include a copy of the License in the document and put the following copyright and license notices just after the title page: Copyright (c) YEAR YOUR NAME. we recommend releasing these examples in parallel under your choice of free software license.. 12. ADDENDUM: How to use this License for your documents To use this License in a document you have written.2 or any later version published by the Free Software Foundation.GNU Free Documentation License option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. to permit their use in free software. no Front-Cover Texts. Permission is granted to copy. distribute and/or modify this document under the terms of the GNU Free Documentation License. If you have Invariant Sections without Cover Texts. and no Back-Cover Texts. and with the Back-Cover Texts being LIST." line with this: with the Invariant Sections being LIST THEIR TITLES. with no Invariant Sections.. Version 1. 54 .
55 . The source can be built into either HTML or PDF format. though building from the source is system dependent. Read without renumeration and dedicated to the programmers of Hire. A build system and special instructions on processing are part of the transparent copy package.com.edu. It is available in transparent copy in the DocBook/XML format at http://samizdat. The original version of this document was written by Robert L. It was originally written in LaTeX. then ported to DocBook.mines. and extension to other formats should be possible and is encouraged.Colophon This work is licensed under the GNU Free Documentation License.
This action might not be possible to undo. Are you sure you want to continue?